H. D. Aiken 1961 - Reason and Conduct

UNIVERSITY
OF FLORIDA
LIBRARIES
/ 2—
/
(\
v
7
Digitized by the Internet Archive
in 2011 with funding from
Lyrasis Members and Sloan Foundation
http://www.archive.org/details/reasonconductnewOOaike
Reason and Conduct
NEW BEARINGS
IN MORAL PHILOSOPHY
REASON
AMD
CONDUCT
New Bearings in Moral Philosophy
& & &
HENRY DAVID AIKEN
NEW YORK : ALFRED -A KNOPF

•
1962
ACKNOWLEDGMENTS
I wish to acknowledge with gratitude the permission of the following publishers to
reprint in this book extracts and articles from the following copyright works con-
trolled by them. The copyright claimant in every case was the person or organiza-
tion who has given permission to reprint the selection.
Constable & Co., Ltd.: from George Santayana, The Sense of Beauty, copyright
1936 for rights outside the U.S. and Canada; copyright 1936 by Charles Scribner's
Sons for U.S. and Canadian rights.
J. M. Dent & Sons, Ltd.: from George Santayana, Winds of Doctrine, copyright
1940.
Harvard Educational Review: for permission to reprint the entire article by Henry
D. Aiken: "Moral Philosophy and Education," Vol. XXV, No. 1, Winter, copyright
1955-
The Journal of Philosophy: for permission to reprint the following entire arti-
cles by Henry D. Aiken: "The Open Society and Its Enemies" (a review), Vol.
XLIV, No. 17, August 14, 1947, copyright 1947, and "The Role of Conventions in
Ethics," Vol. XLIX, No. 6, March 13, 1952, copyright 1952.
The Library of Living Philosophers, Inc.: from Paul A. Schilpp, ed., The Phi-
losophy of George Santayana, Vol. II, copyright 1940.
Philosophical Review: for permission to reprint the following entire articles by
Henry D. Aiken: "Definitions, Factual Premises and Ethical Conclusions," Vol.
LXL, No. 3, July 1952, copyright 1952, and "What Is Value? An Essay in Phil-
osophical Analysis," Vol. LXII, No. 2, April 1953, copyright 1953.
Philosophy and Phenomenological Research: for permission to reprint the fol-
lowing entire articles by Henry D. Aiken: "The Spectrum of Value Predictions,"
Vol. XIV, No. 1, September 1953, copyright 1953, and "The Authority of Moral
Judgments," Vol. XII, No. 4, June 1952, copyright 1952.
Routledge & Kegan Paul, Ltd.: from Karl R. Popper, The Open Society and Its
Enemies, 4th ed., copyright 1962.
Charles Scribner's Sons: from George Santayana, The Sense of Beauty, copyright
1936 for U.S. and Canadian rights; copyright 1936 by Constable & Co., Ltd. for
rights outside the U.S. and Canada.
The University of Chicago Press: for permission to reprint the following entire
articles from Ethics by Henry D. Aiken: "The Levels of Moral Discourse," Vol.
LXII, No. 4, July 1952, copyright 1952; "Moral Reasoning," Vol. LXIV, No. 1,
October 1953, copyright 1953, and "God and Evil: A Study of Some Relations be-
tween Faith and Morals," Vol. LXVIII, No. 2, January 1958, copyright 1958.
The Viking Press, Inc.: from Thorstein Veblen, What Veblen Taught, W. C.
Mitchell, ed., copyright 1945.
L. C. catalog card number: 62-12564
THIS IS A BORZOI BOOK,

PUBLISHED BY ALFRED A. KNOPF, INC.
Copyright ©1962 by Henry David Aiken. All rights reserved. No part of this book
may be reproduced in any form without permission in writing from the publisher,
except by a reviewer who may quote brief passages in a review to be printed in a
magazine or newspaper. Manufactured in the United States of America, and dis-
Random House, Inc. Published
tributed by simultaneously in Toronto, Canada, by
Random House of Canada, Limited.
FIRST EDITION
for Helen
PREFACE
Although the papers selected for inclusion in this volume have

been written at various times and for a variety of occasions
over a period of more than a decade, they are not in the usual
sense occasional works. Much of the serious philosophical
writing in our time has been in the form of the essay. In my
own case, that form is suited precisely both to what I have
hitherto had to say in philosophy and to the way in which I
have wished to say it. I am content to let these papers speak
for themselves, with a minimum of explanation and apology.
Such unity as they possess is the natural unity imposed by a
particular philosophical temperament and style, a character-
istic way of going at problems, and a persistent concern with
a group of problems that appear to me crucial for moral phi-

losophy. Of course, some editorial revision has proved un-
avoidable. Now and then I have discerned an ambiguity or
misleading statement which, upon reflection, has seemed to
me unfaithful to what I remember of my original intention. In
a few cases, I have rearranged, condensed, or expanded dis-
cussions in the interest of greater compression or clarity. But
for the most part I have tried not to second-guess my earlier
opinions or to remove all evidences of my philosophical de-
velopment. In fact, it seems to me that the dialectic of that
development may prove more instructive than any thesis or
antithesis contained therein. By experimenting, I have also
found that it is impossible, even were it desirable, to rewrite
whole essays. For the result is always a new work which
usually says less distinctly a very different thing. And it is
viii Preface
these essays, not some others, that I have thought worthy, if

any of my writings on moral philosophy are so, to be rescued
from the oblivion of periodical publication. I am grateful to
the editors of various philosophical and literary journals for
their past indulgence toward my pieces; I also extend my
thanks for their kind permission to reprint them here.
In arranging these essays, I have decided not to follow a
strict chronological order. Instead, I have divided them into
two main groups. The first represents my own successive
efforts to find answers to the questions of analytical ethics.
The second a selection of critical studies of particular
offers
works that reflect some of the prevailing winds of doctrine in
contemporary moral philosophy and several essays dealing
with a variety of topics, historical and substantive as well as
methodological, which, at one time or another, have engaged
my attention. Taken together, the papers do seem to me to
provide a fairly comprehensive survey of a good many of the
issues with which moral philosophers are now concerned.
However, I most particularly desire not to give the impres-
sion that without really trying I have somehow managed to
construct a new system of moral philosophy. These are "new
bearings," nothing more.
On the whole the perceptive reader will find that the point
of view of my more recent papers is rather more "protestant"
and more 'libertarian" than that of their predecessors. He
will also find, especially in the second group of essays, an in-
creasing preoccupation with certain limiting questions of
moral philosophy and a corresponding broadening of my con-
ception of its scope. Indeed, in the end it has become a ques-
tion for me whether any sharp line at all can be drawn
between moral philosophy and metaphysics or "first philoso-
phy." But even in the earlier papers included in the present
series I always insisted that, especially at its fringes, morality
is necessarily something more, as well as something less, than
a system of socially authoritative rules of action and that no
such system could conceivably prescribe "the whole duty of
man." A decade ago it seemed less important to "make a
thing" of this fact than to insist upon the equally essential
Preface ix
fact,which had temporarily been forgotten, that morality

(like itself) is, through and through, a social phe-
philosophy
nomenon, and that the forms of deliberation and action which
it countenances are not, in particular communities, as open-
textured as proponents of the emotive theory of ethics previ-
ously had suggested. At the time, it needed to be said that
what we ordinarily understand by the moral life is, in routine
situations, subject to routine methods of correction to which
ordinary persons normally submit without question. To a
greater or lesser degree, most moral behavior is standardized
behavior, and the built-in norms which most of us acknowl-
edge as standards of what, in conventional situations, is right
and proper tend at the same time to become de facto rules for
the application of evaluative terms in those situations. Nor can
there be any proper understanding of those levels of moral
reflection which transcend such standardization that does
not begin by noticing that transcendence always begins in re-
sistance or rebellion, and that alienation from the "in-group"
routines to which any individual is conditioned from birth
presupposes that such routines are already in being as de-
terminants of individual conscience and judgment.
This lesson, I take it, has now been sufficiently well learned.
At present there is a danger that contemporary philosophers^
who stress the ordinariness of the ordinary rule-governed lan-
guage of morals, may have forgotten that, at least in our
Western tradition, the "institution" of morality has always
contained within itself the seeds of its own destruction as an
institution. Other philosophers, from Bergson to Popper, have
stressed that our Western moralities are essentially "open."
What this means is that there has gradually been built into
our very concept of morals the possibility of its continual re-
construction. Moreover, owing largely to the profound influ-
ence of great moral philosophers since the time of Socrates
upon the language of morals itself, there has gradually been
included within the compass of the idea of morality the notion
of a philosophical or reflective morality, which, by the nature
of the case, is the notion of a personal morality of spiritually
free men who are not only determined to govern themselves,
x Preface
but also, in turn, determined by the idea or ideal of them-

selves as autonomous, self-governing moral agents. As I have
increasingly come to see, that great philosophical idea is ir-
removable from our whole conception of the moral life. In

brief, what philosophers have called the ordinary language of
morals is there systematically ready for the use of reformers
and revolutionaries as well as of conformists and conven-
tionalists. This, in a culture which is incurably ideological, is
a great boon, for it permits moralists of the right, center, and
left to participate in an endless colloquy about man's spiritual
destiny without finally having either to transcend or neces-
sarily to abuse the language of good and evil.
It remains for me to acknowledge certain obligations and
affinities and to define their limits. That I owe much to the
English philosophers of ordinary language will be apparent

to any informed reader. It is also a matter of record that on
this side of the Atlantic I was one of the first writers on moral
philosophy to make any use of their ideas. What may be less
obvious perhaps is that I have read them for what I could
learn from them, not for what they may have wished to teach
me. From a logical point of view at least, they sometimes write
as if they were revolutionaries who, breaking with everything
that has gone before, are advocating a philosophy to end all
philosophy. Such is not my conception of their work. In my
view, they are in effect rebel loyalists who, after the pro-
longed "Russellian Protectorate" of "scientific philosophy," are
now helping to restore, correct, and deepen the great human-
istic, "common-law" tradition of British philosophy that has
ever sought to preserve a continuity between philosophical

sophistication and common sense and between new ways of
"ideas" and tried ways of doing things. So far as it concerns
me, their revolt is not meant to deny either the importance of
science or the authority of its methods within their own
proper sphere; nor is it meant to assert, as one follower of
Wittgenstein has said, that ordinarily language is the one and
only correct language. Rather is it meant to deny that what is
significant in ordinary utterances can always be translated
without loss of meaning into statements acceptable to scien-
Preface xi
tists as such, and that only when such (unknown) translations
have been made is it possible to determine the truth-value of

ordinary statements. More important, it is meant to deny that
preconceptions on the part of "scientific philosophers" con-
cerning the methods of scientific inquiry are to be identified
with cognition itself and that scientism is anything more than
a rather stupid philosophical prejudice.
Bertrand Russell and his followers pretend that what they
call "linguistic philosophy" is a know-nothing philosophy
which uses ordinary language as a shield to defend its own
ignorance of science and its methods. I believe this is a com-
pletely false issue. The preoccupation of the linguistic philoso-
phers with the meanings of words or concepts represents no
radical departure from their own investigations. Since the
time of Occam, British philosophers of all schools have been
incurably addicted to the study of language. Most of them,
rightly or wrongly,have supposed that a sound philosophy of
language would throw a flood of light upon many traditional
metaphysical, epistemological, and ethical problems that have
hitherto resisted solution. Russell's own best work as a philos-
opher has been concerned with the analysis of concepts, ordi-
nary and otherwise. On this score, the only difference between
him and the ordinary language philosophers is simply that the
latter believe that many of the analyses offered by Russell and
his followers are confused or mistaken, not only in detail, but
in principle. So far as it is rational, this disagreement can be
settled, not by pronunciamentos, but only by careful, piece-
meal study of the uses of particular expressions in the contexts
of their customary employment.
It has fallen to my lot for a good many years to teach British
philosophy to Harvard students, both graduate and under-

graduate. This duty has required me to do a certain amount
of reading in original sources. What I have found there is not
at all what the historians, including Russell, had led me to
expect. To be sure I did find, for example, Berkeley's famous
statement, esse est percipi. I also found many other things,
including Berkeley's nearly forgotten Alciphron, in which
there exist, as plain as day, many of the leading ideas lately
—
xii Preface
celebrated by the followers of Wittgenstein. Something of the

sort may also be said David Hume, whom Rus-
in the case of
sell and the have extolled as the true phil-
logical positivists
osophical Copernicus. What both Berkeley and Hume came
gradually to understand is that the meanings of the concepts
which are the primary concern of the philosopher such con- —
cepts, for example, as truth, beauty, validity, and goodness
can be better understood not simply by adhering more reso-
lutely to the dictum, "Hunt the referent," but only by examin-
ing what men are doing with words when they assert the
truth of particular propositions, when they affirm the validity
of an inference, when they judge the beauty of a poem or the
goodness of an object. This form of analysis, now so wide-
spread among the linguistic philosophers, is brilliantly exem-
plified, for example, in Hume's analysis of promises. In fact it
is exhibited throughout his Treatise with results that are in-

exhaustibly suggestive to the analytical moralist. In my own
case, the reading of Hume, on whose moral philosophy I also
wrote my doctoral dissertation, prepared me well for the ethi-

cal studies of the linguistic philosophers. And it is for this
reason that neither their methods of analysis nor their results
struck me as a bolt from the blue.
I fancy that not all of my English colleagues at Oxford and
Cambridge will be thrilled by these protestations of affection.
For some of them appear to agree with C. D. Broad that
America is simply the place where English philosophies go
after they are dead. From the beginning, of course, American
philosophy has always been profoundly influenced by its Brit-
ish counterpart, and the situation is no different at the present
time. The fact remains that British philosophy, like the lan-
guage with which it has always been so preoccupied, always
undergoes a sea-change in its passage to these States. This is
true in the present instance. I am well aware that I have read
Wittgenstein and Wisdom and Austin, as I also have read
Berkeley and Hume, with glasses ground by the American
pragmatists, Peirce, James, and Dewey. Like William James's
great brother Henry, I too have regarded myself, since my
nonage, as a pragmatist of sorts, and for this reason, it has
Preface xiii
never occurred to me that the study of language could be

separated from the study of the various activities, scientific,
moral, religious, or artistic, of which language forms but the
leading part. But, for the same reason, I have also never sup-
posed for a moment that for the philosopher the study of any
form of words could ever be an end in itself. Philosophy is, or
ought to be, the pursuit of wisdom, and the philosophical in-
terest in language, as in any symbolic form, is therefore al-
ways impure. What the philosopher ultimately seeks, as Peirce
said, is self-control and self-knowledge; he wants to know the
ways of the concepts he employs only in order that he may
use them more successfully in compassing his own ends,
whether material or otherwise. It is precisely this pragmatic
philosophical goal which the English linguistic philosophers
have too often concealed from their readers, if not from them-
selves. As a pragmatist, I am bound to think that every phi-
losopher, analytical or otherwise, is first and last a moralist,
and that his study either of the 'language" of morals or of any
other language is merely preliminary to the conduct of life.
From the pragmatists I learned also that what men call

"reason" is not a Logos written into the very constitution of
being, but an all-too-human regimen or "sentiment," as James
called it, whose claims are neither absolute nor unchangeable.
The underlying insight, again, was Hume's, and it provided
for him the basis of his thoroughgoing critique of rationalism.
Its great philosophical point was also seen by Kant, who de-
voted his own greatest works to the critique of reason in all of
its major deployments. However, both Hume and Kant lived
before the age of the historical consciousness, and it remained

for the great idealists to view reason historically and in-
stitutionally as a system of evolving norms or standards of
validity, correctness, and truth. At the same time, the idealists
intermittently understood that the "objective" standards of
rationalityembodied in such institutions as science, law, and
education can exert no absolute claims upon the minds of men
unless they themselves are prepared, as conscious subjects, to
accept them. Moreover, history does not miraculously make
itself; it is made by men. This means that what Spinoza called
xiv Preface
"the improvement of the understanding" (or reason) is not

something which occurs by chance but in consequence of the
deliberations and decisions of individual persons and groups.
In short, the sentiment of rationality is at once an acquired
and a variable sentiment, whose objective forms are in many

cases the result of proposals which in the first instance can
claim only a "subjective" validity for individuals who may be
willing to adopt them. All this, in one way or another, I have
learned primarily from the German philosophers of the nine-
teenth century, and their influence upon my thought has ac-
cordingly been profound.
One final turn of this screw remains to be made. Many of
the German historicists tended, like Hegel, to become lost in
a maze of cultural history in consequence of which they also
tended to allow the history of reason to double in brass as its
critique. This error lies at the root of the notorious "panlog-

ism" of Hegel which Kierkegaard so violently attacked and
which Nietzsche so brilliantly ridiculed. For Hegel, the his-
tory of philosophy tended to become its very principles. But
the have plausibly argued, in effect, that any
existentialists
"discipline" whose only principles can be its own history is
no discipline at all. They have concluded, and rightly in my
opinion, that there can be no such thing as the principles of
philosophy, and that philosophy must therefore always be a
first-personal activity whose only business, finally, is to dis-
cover what the "self" of the individual philosopher really is,
or, better, is becoming. In short, as Socrates long ago per-
ceived, philosophy at bottom is the search for self-knowledge
and self -revelation. It may be obliged to do many things in
conducting the search. Nor is it true, as some existentialists
seem to think, that the best way of conducting it is simply to
make a decision or else to stare at one's own navel. He who
knows nothing of the world and of history will usually con-
fuse genuine revelations of the self with idle dreams or else
will discover a self that is impoverished, confused, and be-
nighted. Here, it seems to me, Hegel's instincts were sounder
than most of the existentialists who criticize him. I myself be-
lieve, at any rate, that, in Santayana's phrase, it is always "a
Preface xv
long way round to Nirvana," and that in the course of trying

to find out what I myself am and ought to be, I must make
many detours throughout the realms of being.
I am aware, a profound debt to Santayana him-
finally, of
self, whose philosophy James once described as "the perfec-

tion of rottenness." Like that of the pragmatists, whom he
affected to despise, Santayana's is in essence a moral philoso-
phy, and this is true, as it was also true in the case of James
and Peirce, even when he is doing metaphysics and epistem-
ology. Perhaps the greatest value of Santayana's writings for
me is the almost fanatical detachment with which he ex-
amined the various symbolic forms that compose not only the
life of reason but also those supra-rational forms of life in
which the life of reason itself is always incysted. And in that

detachment I have found both a great wisdom and a method.
I suppose that I have returned oftener to Santayana's writings
than to those of any other single philosopher save Hume.

They have nourished me and given me assurance of my own
vocation. Were he the only American philosopher, that fact,
ironically, would suffice to prove that, like Greece and Eng-
land, America has somehow managed to become a mother of
true philosophy.
To my own American contemporaries my debts are so
many and so extensive that it would be futile, as well as in-
vidious, to try to enumerate them all here. However, I cannot
forbear to express my continuing gratitude to a distinguished
teacher, Reginald Arragon of Reed College, from whom I first
learned both the excitement of philosophy and the importance
of understanding something of its history. My debt is only
less great to the writings and to the stimulating conversation
of my friends Raphael Demos, William Earle, William Fran-
kena, John Ladd, Asher Moore, Everett Nelson, Ernest Nagel,
W. V. Quine, Israel Scheffler, Charles Stevenson, Donald
Williams, Morton White, and Paul Ziff. I have been blessed
also by a number of students, in particular Stanley Cavell,
Marshall Cohen, Thompson Clarke, Burton Dreben, Jan
Narveson, Moreland Perkins, and Irving Singer, to whom I
owe more for stimulus and criticism than any of them can
xvi Preface
possibly realize. To have such students is itself sufficient com-

pensation for the often dull and wearing grind of academic
life.Finally, I must once again invoke the dear name of
Ralph Barton Perry, whose memory provides for me an ex-
emplary image of what it means to be a philosopher.
I dedicate this book to my wife, Helen Rowland Aiken,
without whose encouragement the idea of gathering these

essays together would never have seemed a serious matter.
Henry David Aiken

Bedford, Massachusetts
November, iq6i
CONTENTS
PART ONE
i • Moral Philosophy and Education 3
n • The Multiple Roles of the Language of
Conduct 33
in • Definitions, Factual Premises, and Moral
Conclusions 44
rv • Levels of Moral Discourse 65
v • Moral Reasoning 88
vi • The Authority of Moral Judgments 111
vn • Global Conventions in Ethics 128
vin • The Concept of Moral Objectivity 134

ix • God and Evil: A Study of Some Relations
Between Faith and Morals 171
PART TWO
x • Commonsense Ethics and Ordinary
Language 209
xi • A New Defense of Ethical Realism 227
xii • A Revival of Ethical Naturalism 240
xviii Contents
xm • History, Morals, and the Open Society 263

xrv • Utilitarianism and Liberty: John Stuart
Mill's Defense of Freedom 293
xv • George Santayana: Natural Historian of
Symbolic Forms 315
xvi • Philosophical Analysis and the Spiritual Life 349
Index follows page 375

PART ONE
[I]
Moral Philosophy
and Education
CURRENT PHILOSOPHY AND LIBERAL EDUCATION

The advancement of learning is rarely attended by a ready
understanding of its relevance to liberal education. It is not
hard to see why this should be so. On the one side is the
scholar's desire to get without being continu-
on with his job
ally side-tracked by importunate questions about the prac-
tical relevance of his researches; on the other is the
understandable reluctance of educators to range too far from
the highway of received ideas. From the nature of his calling
the educator is bound to rely heavily upon theories and
principles lying somewhat to the rear of current inquiry. The
more radical its departure from tradition, therefore, the less
likely is any new development in a branch of learning to

find its way immediately into the core curriculum. When
first introduced, it is almost necessarily restricted to courses
intended for the specialist. The result is an inevitable, but
not necessarily unhealthy, tension between the concerns of
productive scholarship and the more conservative obliga-
tions of the common educational process. Occasionally, how-
ever, there sets in a more serious dissociation between re-
search and education which can be exceedingly harmful to
the permanent interests of both.
For some time such a situation has been developing in
philosophy. This is doubly unfortunate, since a widespread
4 Reason and Conduct
revival of educational interest in philosophical problems is
occurring precisely at a time in which there has been a

very rapid development in the methods of philosophical an-
alysis that promises results of great value to educated men.
As rarely before, thereis an intense conviction on the part
both of educators and of the general public that a sound

education must include forms of instruction that transcend
the scope of the special sciences. The phenomenal growth of
general education is itself evidence of a conviction, not al-
ways clearly articulated or understood, that only studies
having a philosophical dimension can overcome the cultural
parochialism and the trained incompetence that are thought
to be the by-products of scientific specialization. In prac-
tice this has usually meant a headlong return to the study of
the great philosophical classics, with a corresponding stress
upon the history of philosophy and a tendency to ignore or
even to disparage more recent achievements. The latter, so
far as they are considered at all, tend to be regarded as hor-
rible examples of scholastic logic-chopping that are wholly
divorced from the profounder human needs of our time.
Such phrases as "symbolic logic," "philosophical analysis," and
"logical positivism" conjure up in many academic minds
only images of an outlandish professional jargon and a
recherche symbolism whose meaning is intelligible, if at all,
only to a few godless initiates, and whose point is scarcely
understood even by them. For all the external world knows
or thinks it needs to know ( and unfortunately this includes a
good many teachers of philosophy), such symbol mongering
is of interest only to a few monomaniacs who make a fetish
of something they call the methodology of science and who
thereby forfeit the philosopher's ancient prerogative to speak
and be heard on the primordial problems of man's fate. The
only contemporary philosophical ideas that have received
any really widespread attention outside the seminar are
those impressionistic, semi-literary doctrines, such as ex-
istentialism, which least adequately represent the hard and
sober thinking which has quietly made of our age one of the
few periods of genuine creative advance in philosophy. The
Moral Philosophy and Education 5
result has been that philosophical theories of great rele-

vance to general human enlightenment remain largely un-
known or else are badly misrepresented, while at the same
time, half-cooked ideas, whose very intelligibility is in doubt,
are treated with polite respect and attention. Indeed the very
classics themselves, which should be approached as precious
harbingers of truth rather than as final or official statements
of the views they represent, are set in prejudicial contrast to
the present aims of philosophical research.
The truth is, of course, that any such opposition between
the intellectual concerns of the great thinkers of the past,
such as Aristotle, Locke, or Kant, and those of such origi-
nal contemporaries as Russell, Wittgenstein, and Dewey is
quite illusory. The present interest in problems of method-
ology and in the logical analysis of such concepts as meaning,
truth, and value is merely a continuation of a perennial
philosophical interest in the clarification of basic general ideas
and techniques of inquiry that is older than Plato. If there is
a difference, it lies rather in the fact that contemporary an-

alytical philosophershave found ways of handling these an-
cient questions which are clearer and more precise, and
which, therefore, hold the promise of answers upon which,
for the first time, reasonable men might be expected to
agree. Certainly the contemporary state of philosophy leaves
much to be desired, but its insistence upon clarity, order,
and exactitude is all that has ever distinguished philosophy
as a discipline from common-sense speculation concerning
the organizing concepts by which we live.
It is my purpose in what follows to correct in some measure
the prevailing lack of understanding of the humane import of
analytical philosophy by showing some of the broader im-
plications of recent findings in ethical theory. Moral philos-
ophy in the twentieth century has entered upon a period of
important theoretical advances which, when properly under-
stood, cannot fail to have a more general significance for
reflectivemen. I shall first indicate what some of these de-
velopments are and then say something about their rele-
vance to ordinary human affairs.
Reason and Conduct
SUBSTANTIVE ETHICS AND ANALYTIC ETHICS
Most analytical philosophers are now agreed that a primary-

source of the faltering progress of earlier moral philosophy is
the general failure to distinguish between two very different

sorts of questions which we shall call questions of "sub-
stantive ethics" and "analytic ethics." As will presently be
seen, these questions are not wholly unconnected. But even
to understand how this may be so requires that at the outset
the point of each should be clearly distinguished.
Substantive Ethics. The fundamental question of moral
philosophy has always been "What is the good life?" From
the attempt to answer it comparisons and
arise the endless
reappraisals of specific standards of value and systems of hu-
man conduct. Belonging to this side of the subject are such
hoary but never irrelevant questions as "What, if any, is the
highest good?", "What are the basic principles of right ac-
tion?", and "What are the common rights and responsibilities
of men?" These questions are commonly held to belong to
"normative ethics." So to speak of them, however, is perhaps
question begging, for it already suggests a certain generic
theory as to the nature or function of moral judgments. In
order not to prejudice the discussion in advance I shall
refer to that side of moral philosophy which addresses it-
self to the question how we ought to live as "substantive

ethics." In so doing I wish only to indicate that any other
form of philosophical reflection about morals must take its
rise from the fact that substantive problems of valuation
and obligation already exist. Were it not for them, the ana-
lytical questions with which we shall here be mainly con-
cerned, would have no subject matter.
Now the illusion prevails in certain quarters that the tasks
of substantive ethics are, at bottom, not intellectual but
merely decisional. On such a view analysis can have little or
no bearing upon the problems of the moral life. In a later
portion of this essay I shall try to show how mistaken such a
view is. Meanwhile it must suffice to remark that the possi-
bility of substantive moral philosophy as a discipline pre-

supposes that, at least in some sense, moral ideals are not
beyond the reach of rational reflection and comparison. It
assumes, or seems to assume, that some ideals are not merely
preferred to, but are also more reasonable and hence better
than others. I do not doubt that the interests of substantive
1
ethics are finally practical as well as speculative. But they
would appear to rest upon the conviction that in some rele-
vant sense practice may be informed, and that moral criti-
cism is not just expression of personal taste or feeling. At the
moment, however, I do not wish to stress this point. For pur-
poses of contrast I wish rather to emphasize the point that
although, as I believe, substantive ethics cannot dispense
with rational methods, its aims remain primarily delibera-
tive, and hence are incomplete until they issue forth into
choice, even though the choices with which it is concerned
are at a certain remove from the immediate decisions of daily
life.
Analytic Ethics. The questions of substantive ethics are

no longer denied relevance to the concerns of liberal educa-
tion. On the contrary, it is with them, I suppose, that general
education as such is very largely preoccupied. It is the sec-
ond which marks the special province of
sort of question,
analytical moral theory, that is more frequently thought to
lie apart from the general educational interests of ordinary
persons. Questions of this sort are the following: "What are

the meanings or uses of 'ought,' right,' and Value'?", "What
are the roles of moral reasoning?", "What is the nature of
moral disagreement and of moral justification?", "What is the
relation of factual statements to moral judgments?". By call-
ing such questions "analytic," I wish merely to indicate that
they are concerned in the first instance with "second-level"
problems of meaning and logic rather than with "first-level"
problems of deliberation and choice. It is essential to bear in
1 Speculation is not opposed to practice. Nor is it always directed to
questions concerning what exists rather than what ought to exist. Prob-
lems of conduct often require speculation, not only about matters of
fact, but also about the validity of moral principles themselves.
mind at the outset, however, that interest in analytical

questions is not new; in fact it is nearly as old as moral
philosophy itself, and seems to have arisen spontaneously the
moment that reflection upon moral problems went to funda-
mentals. In the dialogues of Plato, for example, Socrates con-
stantly and insistently asks of his pupils, "What is justice?" or
"What is the nature of goodness?" And from Aristotle onward
questions concerning the "nature" or "essence" of virtue and
obligation and concerning the logic of practical reason have
preoccupied most first-ranking classical moral philosophers.
Confusion of Substantive with Analytic Questions. Until
fairly recently, however, the import of such questions was
not sharply distinguished from those of substantive ethics,
with the result that even some of the greatest philosophers
remained unclear as to the character of the problems to
which they addressed themselves and were in consequence
uncertain as to the sorts of arguments or evidence that might
be relevant to their solution. Until very lately, in fact, the
same tendency to confuse questions of meaning and logic
with substantive ethical issues has plagued many contempo-
rary analysts themselves. Even G. E. Moore, 2 who is perhaps
as responsible as anyone for setting analytical ethics upon
its present course of inquiry, tended to conceal from him-
self the nature of his task by asking his own analytical ques-
tions in a misleading form. He recognized intuitively that
there is a fundamental difference between questions of the
form "What things are good?" and those which ask "What
does 'good' mean?" Yet particularly in his earlier writings he
constantly phrased questions of the latter sort in various con-
fusing ways. By assimilating the question "What is the
meaning or (better) use of 'good'?" to such apparently simi-
lar questions as "What is the property of goodness?" or, more
simply, "What is goodness?" he was led to suppose that by
answering the analytical questions he had put himself, he
would also be able, once and for all, to lay the cornerstone
2 G. E. Moore, Principia Ethica (Cambridge: Cambridge University
Press, 1929).
of a substantive "moral science." How dubious such a sup-

position is will be seen as we proceed.
The Linguistic Character of Analytic Problems. It is es-
sential to understanding of the tasks of analytical ethics to see

that such questions are, at bottom, misleading ways of
all
asking about the meanings or uses of words. They are ob-

jectionable because they tend to conceal from the ques-
tioner the point of his question and so obscure the sorts of
answer that would be relevant to it. First of all, by asking
about the property of goodness, it is assumed that there is
such a property which can be discovered by inspection of
and abstraction from the observable characteristics of good
things. This implies that we may by-pass all linguistic con-
siderations regarding the usage of words and proceed at
once to the important matters of fact to which they sup-
posedly refer. This is a basic error. For until we have ex-
amined the use of "good" in discourse, we cannot know
whether it functions referentially at all. 3 Moreover, even
when words do function referentially, we still cannot de-
termine what they mean by ignoring them and inspecting
their supposed referents. It is a wise referent indeed that
knows what it is called, and a wise observer who can tell by
what color is named by the word "redness."
"inspection"
However, once we see that problems of definition and
—
meaning in short, problems of analysis can only be solved —
by close study of the uses of words, we gain at once a pur-
chase upon the nature of our task. In this way we are able
for the time to delimit the range of relevant disagree-
first
ment and hence seriously to hope for a settlement of the

endless disputes that for centuries have divided the various
schools of ethics.
The Human Relevance of Linguistic Analysis. At first
3 I.e., whether there is in fact some objective quality or relation

which it purports to designate; whether its purpose is, as a matter
of fact, to designate, describe, or inform at all. The terms "refer-
ential," "descriptive," and "designative" are used interchangeably
throughout.
io Reason and Conduct
sight problems of meaning may appear to have little rele-

vance to the substantive problems of human conduct. The
question of how we are to use such words as "ought" and
"good" may be of theoretical interest to the logician or seman-
ticist, but the answer to it, as the saying goes, "butters no
bread." On this point not merely the opponents of analytical

philosophy but also a good many analysts have been agreed.
And it must be confessed that there is some initial plausibility
to the argument that analytic philosophy cannot have it both
ways: it cannot claim the complete irrelevance of substantive
or factual considerations to analytical questions without at
the same time admitting the irrelevance of its own verbal
and logical concerns to substantive or factual questions.
Such views I take to be ill-considered, and I shall try
subsequently to show that philosophical analysis is not
the academic concern of a few methodolatrists and word-
mongers who have thereby turned their backs upon the
pursuit of wisdom, but an indispensable means to clear
thinking and relevant argument in the conduct of practi-
cal affairs. Ways of life are articulated through words. If we
misunderstand the ways of the words through which they
are expressed, so also will our grasp of the ways of life them-
selves be faltering and confused. If we radically misconceive
the roles of moral judgment and the nature of moral reason-
ing, so also will we stumble in the use of moral discourse.
And if we are unclear as to the meanings of the great terms
that lay down the practical ends-in-view which are to guide
our lives, there can be no clarity in the ends themselves
and no settled direction to our conduct. In no domain of
human activity is more linguistic confusion and fal-
there
lacious thinking than in the domain of morality, and in none
therefore can a greater benefit be hoped for from the study
of the language in which the activity is clothed.
In order to give body to these remarks, however, it will first
be necessary to follow for a bit the progress of analytical
moral theory in recent decades. In this way we will be better
able to see just how recent findings in this domain may be of
use in the conduct of life.
THE DEVELOPMENT OF RECENT ANALYTIC ETHICS

Descriptivism and Its Defects. Most earlier proponents
of analytical philosophy tended to take for granted that the
primary function of language is the communication of
knowledge or information about matters of fact. They as-
sumed as a matter of course, that the task of the analyst is to
fixmore clearly the supposed designata or referents of the
terms with which he is concerned. Such an assumption can
do relatively little harm so long as the words under analysis
—
happen to be used descriptively so long, that is to say, as
the sentences in which they occur are employed to character-
ize and predict events. It is an egregious assumption, which
simply blocks the path of inquiry, when the role of expres-
sions is not essentially informative. As we shall see, it is
fatal to the understanding of ethical terms whose primary
uses are commendatory and prescriptive, and which only
secondarily, and in certain special circumstances, serve
incidentally to characterize the things of which they are pred-
icated. And in humanities generally, it at once results in
endless and futile disputation, and fortifies, where it does not
itself produce, profound confusions of value that have been
endemic throughout the history of our intellectualistic and

rationalistic culture. When what you want to know is the
"cash value" of such words as "yellow" or "buffalo," the rule
"Hunt the referent!" seems sound enough advice, although
even in such cases a flat-footed use of it can impede rather
than advance inquiry. But where words are not used refer-
entially, and the utterances in which they occur are not in-
tended to express anything which "corresponds to the facts,"
the rule actually produces unclarity. By implication, more-
over, it has the disastrous effect of creating the impression
that whena referent cannot be found the word in question is
senseless and its use the business only of deceivers and
fools. Such an impression is inimical to the interests of litera-
ture and religion as well as morals; worse still, perhaps, it
endangers fruitful theory formation in science itself.
The Search for a Science of Morals: Intuitionists and

Naturalists.The referential dogma, as it may be called,
dominated nearly all of the work which was done in analyti-
cal moral theory during the first decades of the present cen-
tury. Even a cursory examination shows that it controls the
direction of the analyses both of such influential "intuition-
ists" as G. E. Moore, 4 W. D. Ross, 5 and C. D. Broad 6 in
England and of the more influential American "naturalists"
such as John Dewey 7 and R. B. Perry. 8 Under its influence,
all of these important writers conceived the primary aim of
moral philosophy to be that of reducing ethics to a science.

They believed that only by means of a definition or clarifica-
tion of the referent of the "primary" ethical terms, such as
"good" or "value," would we be able to state precisely the
conditions under which moral or value judgments, like other
types of "scientific" statement, might be verified. Many of
them by defining such other ethical terms
also believed that,
as "ought" and "right" in relation to the primary terms, it
could be shown that the terms employed in various forms
of valuation all have a common referent. In this way they
sought to reduce or eliminate altogether the necessity for
appeals to special acts of "insight" or "intuition" either in de-
termining the meaning of ethical terms or in certifying the
judgments in which they occur.
Certainly not all of these writers found precisely what
,. they were looking for. Some of them, like Ross, came to
i despair of reducing "right" to "good" or of defining "obliga-
!
tion" in relation to "value." Some, like Moore, were also forced
to conclude that "good" is not an empirical concept at all. Still,
4 Moore, op. cit.

5 W. D. Ross, The Right and the Good (Oxford: The Clarendon
Press, 1930).
6 C. D. Broad, Five Types of Ethical Theory (New York: Harcourt,
Brace, 1930).
7 John Dewey, The Logical Conditions
of a Scientific Treatment of
Morality ( Chicago University of Chicago Press, 1903 ) Dewey, Theory
: ;
of Valuation (Chicago: University of Chicago Press, 1939); John Dewey

and J. N. Tufts, Ethics, rev. ed. (New York: Henry Holt, 1932).
8 R. B. Perry, The General Theory
of Value (New York: Longmans,
Green, 1926).
all them hoped that the hitherto insolvable disagreements

of
in ethics would be capable of a rational resolution if only
the methods employed in the validation of moral judg-
ments could be shown to have the same essential logical
structure as that which has enabled the exact sciences to
move into the thoroughfare of cumulative knowledge. For
intuitionists and naturalists alike, this meant hunting and
finding the common referents (objective qualities and rela-
tions) that are presumably designated by ethical terms.
Thus, for example, Moore, in the opening pages of Principia
Ethica, contrasts the hit or miss affair of ordinary "per-
sonal advice" and "exhortation" with the interest of "scien-
tific which seeks to discover, amidst the apparently
ethics"
random assortment of things which men call "good" or "de-
sirable," that common, objective quality through which alone
we may truly know what things are really good. Subse-
quently he hoped to discover those objective principles of
right action which would supply the objective grounds of
valid moral judgment, thus removing from scientific moral
deliberation that element of prejudice and vague surmise
which characterizes the moral opinions of ordinary men.
Now superficially no ethical theory lies at a further remove
from Moore's than that of John Dewey. Moore was an intui-
tionist whose loyalty to understanding and truth is deeper
than any loyalty he might have to any particular method
such as that of empirical science. For Dewey the two com-
mitments are identical, and he remained throughout his later
life an intransigent naturalist and experimentalist. What
science could not enable him to say, he did not care to dis-
cuss. Yet despite basic epistemological differences in their
respective analyses of ethical terms, Dewey's fundamental
view of the task of moral theory differs little from that of
Moore. The title of one of his most important and influential
works in ethics, The Logical Conditions of a Scientific
Treatment of Morality, itself shows that, like Moore, he took
it for granted that the task of philosophical analysis was to
show in detail how moral discourse is reducible to the status
of a science. Numerous other examples might be given to
show the same underlying assumption that the language of

morals is or ought to be an objective, scientific language,
and that morality itself is or can be rendered intellectually
respectable only by specifying the intersubjective verifying
conditions of judgments of value and right action.
Beyond Descriptivism: Moore's Critique of Naturalists.
Gradually, however, a very different view concerning the use
of ethical terms and the role of moral judgments has
emerged. By a curious irony, the rise of this conception was
due in large part to conclusions implicit in Moore's own
acute analyses of ethical terms. Indeed one of the dramatic
turning points in the history of moral theory occurs when,
despite his own professed intention to reduce ethics to a
science, Moore's own arguments convinced him that any at-
tempt to define moral notions empirically is bound to fail. 9
As stated by Moore, these arguments were less conclusive
than he supposed. Yet they served to convince a whole
generation of moral philosophers of the essential wrong-
headedness of the whole naturalistic program in ethics. In
essence, the argument went something like this: Of any
empirical characteristic which may be designated by such
words as "desire," "pleasure," "satisfaction," or "adjustment"
(the words, that is to say, in terms of which naturalists have
\ perhaps most frequently defined the terms "good" or
"value"), one may always seriously ask, "But, after all, is it
really good?" x This eternally "open question," Moore be-
\lieved, shows that the property of goodness or value cannot
be identified with any "natural" characteristic, and that the
/ word "good," which designates this property, can never have
ythe same meaning as any expression designating a "natural"

characteristic. In fact, Moore held that all attempts what-
ever to define ethical terms are at bottom guilty of the same
"naturalistic fallacy" of confusing questions concerning the
meaning of "good" with substantive problems as to the
9
G. E. Moore, op. cit.
1
Moore's question was intended to show the futility not only of
empirical, but also of "metaphysical" definitions of goodness.
He concluded, in a way which

things which in fact are good. 2
is understandable given the premises of his analysis, that
"good" must be assumed to designate a simple, unanalyz-
able,and ( whatever that means ) "non-natural" quality whose
apprehension is entirely beyond the reach of our ordinary
empirical faculties. It is for this reason that Moore is usually
classified as an "int uitionist ," even though that designation
may appear to suggest a kind of philosopher very different
from the common-sensical, "Let's-have-no-nonsense" sort
of thinker Moore professed to be —
and was. But Moore was
no mystic or romantic advocate of private intuitions into
what was never seen on land or sea. It was argument and
logical analysis that convinced him that goodness is a non-
natural quality, not insight which he might claim as a
prophet or a seer.
Beyond Descriptivism: Ogden and Richards' Emotive
Theory of Ethics. Moore's conclusion was plainly unaccept-
able to empirically-minded philosophers, even when they
were impressed by the force of his arguments. To many of
them, therefore, his analysis seemed to imply merely the
failure of the program of "moral science." Negatively, his
insistence upon the indefinability of "good" and upon the
irreducible logical difference between "natural" and "moral"
facts suggested that "moral facts" might not be facts at all.
Evidently what was wanted was a more flexible conception
of the roles of language that was not committed in advance
to the dogma that in all significant uses words function es-
sentially as signs. Such a conception was first introduced in
recent times by C. K. Ogden and I. A. Richards, 3 the pioneers
of the now flourishing discipline of semantics. In effect,
Moore was compelled realm of "non-natural"
to invent a
entities in order to have something for his indefinable
ethical terms to refer to. Ogden and Richards, on the other
2 These arguments and their merits are discussed in more detail in

later essays.
3
C. K. Ogden and I. A. Richards, The Meaning of Meaning ( New
York: Harcourt, Brace, 1938).
hand, rejected altogether Moore's underlying assumptions

about the meanings of words and so were free to draw an
entirely different conclusion from Moore's own arguments.
Such words as "good," they contended, are non-designative.
They serve, not as names, but as emotive expressions of atti-
tude and as imperative prods to action. Accordingly the ut-
terances in which such words occur are not statements which
might be verified as true or false. Even to raise the question
as to the method of their verification involves at the outset
a complete misconception of their linguistic function.
Beyond Descriptivism: The Logical Positwist' Approach to
Ethics. In a very different quarter this view received sup-
port from a new generation of philosophers who were un-
compromisingly empirical in their approach to the method of
human knowledge, but were more impressed than the older
empiricists had been with the potentialities of the new logic
as a tool for the analysis of propositions and for the con-
struction of amore exact or "ideal" language for the forma-
tion and transformation of scientific statements. 4 They ac-
cepted the directive that all "meaningful" statements must
be capable in principle of reformulation in terms of an ideal
language of science through which their verifying conditions
would be more precisely expressed. At first, they contended
that the method of their verification is literally constitutive
of the meaning of all significant propositions, although in re-
cent years they have been compelled to abandon so strin-
gent a condition of "descriptive" meaning. It may appear
from this that the "logical positivists" or "logical empiricists"
as they alternatively called themselves, were still subtly un-
der the spell of descriptivism, and so, indeed, they were.
Hence it might be supposed that they would follow the lead
of the naturalists in trying to reconstruct the language of
morals or valuation in such a way that it might be shown
to conform to the requirements of "meaningful" discourse
as they conceived it. Here, however, education took a hand.
Nearly all of the original group of logical positivists were
4 The work of I. A. Richards seems to have been largely independent
of this movement, though it is clearly related to it.
educated in Austria and Germany where the idealist tradition

was then still the dominant philosophy. Since Kant, the ad-
herents of this tradition simply took it for granted that a
radical bifurcation must be made between the logical and
epistemological foundations of morality and those of empiri-
cal science. The positivists in effect took this bifurcation for
granted, and so were not tempted to follow the lead of the
American naturalists in trying to formulate a science of
morality. But because of their commitment to the ideal of
the "unity of science," they could not accept the idealists'
view of morals as a form of Geisteswissenschaft. If there is
no empirical property of goodness or value, so they main-
tained, there is no such property at all; if moral discourse
does not conform to the canons of the scientific method, then
morality is simply not a form of knowledge. 5 They concluded,
6
therefore, that ethical concepts are "pseudo-propositions."
Their analysis of moral discourse must be also viewed,
however, in another way as an indirect corollary of an ap-
proach to the analysis of signs which regards ordinary
language as an inadequate and logically imprecise vehicle
of communication. Virtually no philosophers at this stage
questioned that words are signs. The task of analysis, as the
positivists conceived it, is not somuch
to determine the
meanings of ordinary expressions, but to replace such ex-
pressions with a more exact or "ideal" language of science.
Now as they are formulated in ordinary language, ethical
sentences are grammatically indistinguishable from other sen-
tences that express bona fide factual propositions. Because
they are expressed in the indicative mood, as are ordinary
factual statements, it appears at first that the predicate
terms of such sentences are used to assign a corresponding

property to the things designated by their subject terms.
Logically, however, ethical sentences are indistinguishable
from imperatives such as "Shut the door" or "Keep off of the
5 See Rudolf Carnap, Philosophy and Logical Syntax (London:

Kegan, Paul, Tench, Trubner, 1935).
6 The same argument, of course, could be used to show that the
concepts of pure mathematics and logic are also pseudo-concepts.
grass." It is indeed precisely this sort of misleading and con-

fusing feature of ordinary language which, according to the
positivists, renders it so poor a medium
and for clear thinking
must be
exact statement; for this reason, so they contend, it
replaced by a correct language which would contain no

meaningless terms and would permit no pseudo-proposi-
tions. 7
Since, then, there no way of specifying verifying condi-
is
tions for ethical judgments, there is no way of showing any
party to an ethical disagreement to be in error. From a logi-

cal point of view, therefore, ethical disagreement is entirely
pointless. But in that case what can be the sense of moral
discourse? It may be nonsense. It is at any rate a very potent
sort of nonsense in which empirically-minded philosophers

and scientists also indulge themselves unofficially when
they are not at work. If value judgments are without mean-
ing, this would appear to be the proof perfect that man is the
irrational animal, since he ostracizes those whom he pro-
nounces "immoral" and is frequently quite prepared to sacri-
fice even his life for the sake of a pseudo-concept. The solu-
tion to this paradox was provided by the distinction, first
formulated by Ogden and Richards, 8 between two radically
different kinds of "meaning." One of them, which is charac-
teristic of the language of science, is "cognitive meaning" or
"descriptive meaning"; the other, which is exemplified in

art, religion, and morality is "emotive meaning." The former
is designative, and the statements involving it may be true
or false; the latter is non-designative and the sentences in-

volving it are essentially expressive and incitive. 9
Problems of Emotive Theories of Ethics. Unfortunately,
almost as many problems as it at first
this distinction created
appeared to solve. In the first place, what in the world is
emotive meaning, and how is it possible? Many critics, such
7 This view was perhaps most clearly stated by A.
J.
Ayer, Language,
Truth and Logic (London: V. Gollancz, Ltd., 1936).
8 Op. cit.
9 This approach was carried to its greatest achievement in the work
of Charles Stevenson, Ethics and Language (New Haven: Yale Univer-

sity Press, 1944).
as Max Black, 1 have questioned whether there can be such a

"mode" of meaning at all; others have doubted whether the
emotive language can be accounted for in the ab-
effects of
sence of any element of cognitive meaning; some have also
argued that it is merely a by-product of cognitive or descrip-
tive meaning. But even if we waive these difficulties, there
are other equally serious problems involved in the use of the
notion of "emotive meaning" as a tool for analysis. If, for ex-
ample, morality, poetry, and religion are all lumped together
as "emotive," how is it possible to distinguish between them?
Yet anyone who is actively concerned with these forms of dis-
course cannot believe that they have the same kind of effect
or are employed in the service of the same kind of end. In a
word, is not the term "emotive" simply another more pre-
tentious, pseudo-technical way of saying "unscientific" which
therefore tells us nothing positive whatever about the forms
of words to which it is applied?
Were there space, even graver objections to the concept
of emotive meaning might be
raised. Nevertheless, it must be
insisted that even this crude, vastly oversimplified classifica-
tion of the meanings of words was a great step forward. Just
because it was so crude and so vague, it has been a goad to
further research and to subtler, more complex theories of
language. And, once for all, it served to destroy the stultify-
ing and pernicious illusion that language is merely an in-
strument of scientific thought. Under it, the question-begging
rule "Hunt the referent!" gradually gave way to a more flexi-
ble and more realistic approach to the variable functions of
symbols.
The Plural Uses of Ordinary Language. More recently,
under the impact of the teaching and later writings of Lud-
wig Wittgenstein, 2 there has arisen within analytical philoso-
phy a new movement 3 which, among other things, has helped
1 Max Black, Language and Philosophy (Ithaca: Cornell University
Press, 1949), Ch. 9.
2 Ludwig Wittgenstein, Philosophical Investigations (Oxford: Baisell
Blackwell, 1953).
3 This school is most clearly and persuasively represented by the
ethical writings of S. E. Toulmin, An Examination of the Place of

to make good the defects and limitations inherent in the posi-

tivisticapproach to the analysis of language. Its proponents
begin with the premise that ordinary language is multi-
functional. They are concerned to determine, by careful
study of particular forms of words, just what the various
positive roles and uses of ordinary language may be. Their
assumption is that, when properly used, ordinary expressions
are never meaningless and that the normal use of words in
ordinary language is never without point. The task of analysis
is thus not to replace ordinary language, but to determine its
uses, and through this to illuminate the wider practices of

which such uses are a part. This school, therefore, radically
rejects the contention of Bertrand Russell that ordinary
language is a crude repository of outworn metaphysics which
cannot be trusted for the clear articulation and communica-
tion of ideas. Like any language, ordinary language is a
tool; and when correctly employed, it is a sufficiently pre-
cise and flexible tool for the ordinary conduct of human
affairs. In practice it is not misleading. It is only when philos-
ophers ignore the contexts of its normal uses and applications
that it proves misleading. Thus, for example, it is only when
we ask what goodness or obligation and by itself, that
is, in
we lose our logical bearings and are tempted to adopt
wrong-headed models for the analysis of ethical terms.
Viewed in context, moral discourse is not a crude, pre-
scientific form of speech which must be reconstituted so as to
render it more amenable to the requirements of a "moral
science." Such a science is not even a possibility, and any
such reconstruction, were it to prevail, would in effect mean
the supercession of the institution of morality altogether. But
neither is the language of morals a merely expressive lan-
guage for the venting and inciting of private feeling or

emotion. Were it to become such, in consequence of insensi-
Reason in Ethics (Cambridge: Cambridge University Press, 1950). As

the reader will presently discover, the affinities of the present writer to
this school are very close, although there are essential differences which
should also be noted.
tive philosophical tampering, the general social consequences

would be incalculable. All models for the analysis and in-
terpretation of moral discourse provide, at best, only par-
tially relevant analogies which are inevitably misleading and
invariably subvert the aims of understanding when taken
more seriously as bases for "logical reconstruction."
The Language of Conduct: Organization of Social Be-
havior. The philosophers who hold this view accept the
thesis that the language of conduct is a practical and dy-
namic mode of speech whose primary role is the organiza-
tion and control of behavior rather than the description or
prediction of matters of fact. So far they agree with the emo-
tivists. Otherwise, however, the two approaches have little in
common; for the new approach takes seriously, as the emo-

tive theory does not, the fundamental distinction between
moral judgments, which profess a certain objectivity and im-
personality, and mere expressions of taste or interest which
neither have nor claim to have any inter-personal "author-
ity" over the judgments or conduct of others. It takes seri-
ously, also, the normal assumption that moral reasoning is
not just a way of "irrigating" someone's attitudes (although
it may be this also), but an impersonal rule-governed mode
of justification to which the concept of validity may be

relevantly applied. And in general the new approach regards y
moral discourse from an institutional viewpoint as a com-
plex, if informal, machinery of regularized social adjustment
and adaptation. Its adherents accept the imputation that
there is a point in speaking of arguments as "irrational" and
of decisions as "arbitrary"; and it takes seriously the pre-
vailing assumption that in speaking of a supporting reason to
a moral conclusion as a "good" or "bad" reason, we are ex-
pressing something more than our own first-personal ap-
proval or disapproval of it. The task of the moral philoso-
pher is not to try to show what practical reasoning would be
or ought to be if it were to conform to certain rules of in-
ference which the formal logician or methodologist may pre-
fer to regard as "valid." Rather it is to elucidate the patterns
of deliberation and justification which, as they stand, de-

termine what we mean in speaking of a process of moral
argumentation as "invalid" or "irrational."
The Language of Conduct: Rule-Governed and Im-
personal. In brief, the new approach takes seriously com-
mon-sense appearances in the light of which the characteriza-
tions of moral discourse offered by the descriptivists and the
emotivists alike must inevitably seem paradoxical. Morality
could not become a science without a radical reconstruction
of the very uses of such terms as "good" and "ought"; but
such a reconstitution would no longer enable us to say (or
do ) "the ethical thing." It would be better, because less mis-
leading, if the descriptivists really wish to transform the
language of morals into a descriptive, scientific form of
speech, simply to say with Nietzsche that we should go be-
yond "good" and "evil" altogether. Such a view is worth con-
sidering; but it should not be allowed to masquerade para-
doxically as a theory of morals. On the other hand, the terms
of moral discourse are, in use, not such wildly "open-
textured" expressions of emotion as the emotivists contend.
Governing the use of ethical terms, in the various distinctive
contexts in which they occur, are rules of application that
prescribe themanner in which ordinary substantive prescrip-
tions and commendations are to be made. Such rules, how-
ever vague or flexible, still set limits to the sort of judgment
we are prepared to countenance as "ethical." Each time we
apply moral rules we are not simply "venting" our own pass-
ing sentiments or wishes; rather, we are in such a case in-
voking an impersonal linguistic ritual which serves to keep
practical deliberation and disagreement within certain so-
cially acceptable bounds.
The point, then, is not that we do not express our own
attitudes in using the language of morals —on the contrary.
If, as we have suggested, the language of morals is an effec-
tive social device for the regulation of human relations, it
would be exceedingly odd if moral judgments did not re-
flect certain basic social attitudes of the speaker. But moral
judgments reflect our attitudes as social animals precisely be-
cause the latter have themselves been largely formed by

habituation to the rules which govern the application of
ethical terms. To parody Marx, it is not so much the individ-
ual conscience that determines the application of ethical
terms, as it is the standard application of the terms which
determines the conscience of the individual moral judge or
agent. In short, it is through the constant employment of the
language of morals that the individual is perforce obliged
continually to re-enact certain impersonal social roles which
willy-nilly render him, or at any rate his judgment, a
"voice" or "conscience" of the social group. And when, for
whatever reason, the individual misuses or misapplies the
terms of moral discourse, he is automatically subject to the
same forms of verbal rebuke which, in other domains, are
directed against those who will not "talk sense" or "listen to
reason."
These suggestions will be further explored in other essays
in this volume. But enough has perhaps been said to indicate
the increasing sensitivity and maturity of recent philosophi-
cal analyses of morals. What should be emphasized is that
the errors to which philosophical analysis are subject, like
those that occur within science itself, tend to be self-
correcting, and that out of the successive misconceptions
and exaggerations which we have had occasion to observe
there is gradually emerging a more and more adequate grasp
of an exceedingly complex symbolic process. 4 Certainly, the
end is not yet. But enough is already understood concerning
the language of conduct to be of use in the education of
ordinary men. What remains to be shown is the relevance of
these findings to general human enlightenment and to the
conduct of life.
4 For example, as I now believe, the remarks made on the preceding

page concerning moral rules are in need of radical qualification and
refinement.
THE PRACTICAL RELEVANCE OF ANALYTIC ETHICS

As I believe, there are two important respects in which this
analytical study of morals may be relevant to the practical
concerns of men. One of these pertains to the findings them-
selves, the other to the method of analysis employed in
reaching these findings.
The Language of Conduct and the Life of the Commu-
nity. Now if, as we have maintained, the language of morals
is a dynamic language which at the same time is subject to
relatively definite rules of application in various contexts,
there follows at once a corollary which is pertinent to the
present purpose. That is to say, the patterns of moral dis-
course, including the prevailing forms of commendation,
prescription, and justification, provide a kind of mirror of the
prevailing patterns of interpersonal relation and hence of
the underlying way of life of the community. To the extent,
also, that the individual himself is not deviant, the analy-
sis of moral discourse will also provide a sort of map of his
character as a social being. Any ambiguities, vaguenesses, or
inconsistencies in the one will at the same time reflect con-
fusions, indecisions, and tensions in the other.For example,
where incompatible analyses arise from an inability to accept
common instances and counter-instances in trying to deter-
mine the range of application of ethical terms, this will at
once suggest, not a failure of analysis on the part of someone
or other, but rather a lack of homogeneity in the moral atti-
tudes of the social group. And conversely, where analysis is
not stalled because of such disagreements, there is indicated
the presence of a common
system of moral habits. Again, a
great deal of interest in defining and redefining ethical terms
will tend to reflect a lack of clarity or consistency in inter-
personal relations or else an inadequacy of communal moral
standards and practices to the satisfaction of underlying hu-
man wants or drives. 5
5 These remarks are further developed in essay three, "Definitions,
Factual Premises, and Ethical Conclusions."

All this being so, it would appear evident that anyone who
seeks to know himself as a moral being will perforce be
obliged to reflect with some care upon the nature of moral
discourse. And if, as most of us profess to believe, a grasp of
the communal way of life is essential to intelligent par-
ticipation in the guidance and control of human affairs,
then the study of the language of conduct must be a matter

of general human concern.
Correcting Popular Misconceptions. In the second place,
it is only through analysis of moral discourse that one gradu-
ally becomes aware of its limits. Such understanding is all
the more important precisely because, unlike mathematical
logic or the theory of induction, ethical theory, in one crude
form or another, is already part of the public domain. As such
it is constantly, if illicitly, used to bolster or lend prestige to
various substantive prohibitions or demands. As any teacher
of ethics soon discovers a great part of the difficulty of in-
struction arises from the fact that most of his pupils already
hold theories —or prejudices —of their own which they have
acquired in the home or school or church. I mean by this not
merely the obvious fact that most persons, by the time they
have entered college, have well-established moral habits and
attitudes but rather that they have fairly well defined views
concerning the meanings of ethical terms and the nature of
moral principles. To one student, so-called Protagorean rela-
tivism seems virtually self-evident; to another, brought up
perhaps in a parochial school, the view that moral "truths"
are laws of nature or divine commandments needs little ar-
gument; to still another, morality itself is a form of prejudice
to be replaced by hygiene and social engineering. In short,
there abound in the popular consciousness a great many
pre-analytic theories of morals, nearly all of which, unfortu-
nately, involve profound misconceptions as to both the char-
acter of moral judgment and the possibilities of moral justifi-
cation. And these, unhappily, deeply interpenetrate the
whole moral consciousness of those afflicted with them.
In what follows, I wish briefly to indicate how the findings
of recent analysis may serve to correct some of these miscon-
z6 Reason and Conduct
ceptions and so to clarify the deliberative practices of the

layman.
For example, ifone clearly grasps the essentially practical
or regulative role of value judgments generally and of the
forms of reasoning intended to support them, one comes
finally to understand, as one can in no other way, why it is
that increased knowledge of matters of fact does not and
cannot compel common agreement at the moral or ideologi-
cal level. 6 In this way one may be fortified against a pecul-
iarly prevalent and virulent sort of disillusionment either
with morals or with science or with both which afflicts those
who have been led to suppose that there is a necessary
connection between knowledge and virtue. If one also un-
derstands the fact that our moral judgments and hence our
moral practice are fashioned in the light of socially condi-
tioned patterns of commendation and prescription, one is far
less likely to misconstrue or to object in the wrong way to
the moral faults or failures of "insight" of those who system-
atically disagree with us. And one will see thereby that the
remedy for such faults and failures is not to be looked for
in statements of principle or in appeals to "right reason"
which beg the very point at issue. "Self-evidence" is nearly
always a sign, not of god-given truth, but of thorough habitu-
ation. By properly learning this lesson through the study of
the logic of moral justification and persuasion, one is thereby
enabled to comprehend a matter of the greatest importance
for intelligent cultural relations, namely, that the moral prin-
ciples of one's own people, however beneficent, are not due
to profounder insight into the metaphysical structure of
the cosmos, but rather, for the most part, to a more fortu-
nate cultural tradition. Conversely one may see that what
we conceive to be moral obtuseness or bigotry is due, as a
rule, not so much to stupidity or ignorance as to a less
benignant human environment. In this way one may also
gain a sense of the precariousness of any moral order and of
6 For a fuller discussion of the concept of ideology and its relation

to morals and to philosophy itself, see my The Age of Ideology (New
York: Houghton Mifflin Co., 1957), esp. Chapter I.
the preciousness of those social and cultural circumstances

in which alone a "reasonable" ethical system can grow and
flourish.
With all this comes a gradual awareness of why it is that
the basic problems of community and world order can never
be resolved by direct religious or political propaganda, but
only by a process of slow re-education. Such understanding
is doubtless sobering; but it may protect the individual and
the society against the self-destroying disillusionment which

attends the inevitable failures of short-term moral crusades.
At the same time it might help to save us from the now
fashionable varieties of irrationalism which ascribe men's
spiritual limitations to original sin and which make a fetish of
human weakness or depravity without shedding any light
whatever upon its causes or the means of its removal.
The Ultimacy of Conscience. Possibly the greatest practi-
cal insight which the study of moral philosophy can impart is
understanding of the fact, which Kant long ago expressed in
another form, that there is no way of "deriving" moral princi-
ples from anything beyond the consciences of men, and
hence that there is no way whatever of guaranteeing the
validity of any moral code by trying to "ground" it in science,
—
or the nature of the universe or in the will of God. Or, bet-
ter, because less misleading, one may see that any such
"grounding" is, merely another symbolic device

in practice,
for articulating and regulating human attitudes which can
be used effectively only among those who already have been
practically conditioned to respond to it. The masks of con-
science are many; the important thing is to understand them
for what they are. Maturity comes when we finally realize
that, however we may disguise the fact by the symbols by
means of which we conduct our communal affairs, every-
thing in this domain is finally "up to us." Then we see that
ours is the only responsibility, and that it is our own con-
stitutional loyalty to principles of liberty, justice, and wel-
fare that can alone provide a relevant practical ground for the
rights and duties which, as we believe, should be the prerog-
atives of every man.
This is altogether a matter of understanding what one can

and cannot do with the language of conduct. The great
trouble with dogmatic theories of moral obligation which in-
sist upon referring it to "natural law" or "divine command-
ment" is precisely that they prevent those who accept

them from seeing why they find such doctrines attractive or
"true," and, conversely, why others do not. By helping to
remove such forms of parochial self-deception and hence of
eventual disillusionment, ethical theory may make a very
considerable contribution to the advancement of common
practical understanding.
In the preceding remarks I have tried to suggest a few of
the ways in which philosophical study of moral discourse
may have a beneficial effect upon our grasp of one important
aspect of human relations. I wish to show
Before concluding
also that the very methods employed by the analyst in try-
ing to penetrate the thicket of human conduct are also useful
and even necessary within the more concrete domain of
substantive ethics itself.
Clarifying Our Fundamental Goals in Life. Now it has

been frequently maintained, and even today most philoso-
phers would appear to hold, that the only practical rele-
vance of reason in the conduct of life lies in its power to dis-
cover possible causes or means to ends upon which we have
already decided. Since Aristotle many philosophers have
argued that rational deliberation is and can be concerned
only with the ways of achieving ends, but not with the
formation of the ends themselves. This, I take it, is the pri-
mary point of Hume's paradoxical remark that reason is and
ought to be the slave of the passions. Such a view clearly
implies that at bottom the formation of a decision is a non-
rational or supra-rational affair. (Those who have held it, in-
cidentally, have not always seen that if this is so, it applies just
as much at the level of means or proximate goals as at the
level of so-called ends or ultimate goals.) Now undoubtedly
there is an important grain of truth in such a view, al-
though it is misleadingly expressed. This grain of truth is
simply the logical point that a moral judgment, whether
commendation, prescription, or imperative, cannot be logi-\

cally derived from a theoretical or factual truth alone. Be- V
cause it has been misleadingly formulated the theory has
given rise to the notion that all we have to do or can do in
reaching a decision is to let nature take its course, to wait,
that is to say, until the strongest of our conflicting interests
happens to manifest itself. Despite its patent irrationalistic
implications, such a view curiously involves an extremely in-
tellectualistic conception of purposive behavior which is con-
tradicted both by common sense and by contemporary psy-
chological theory. For it evidently assumes that although we
may not know in advance how to get them, we all do know
at the outset what our ultimate goals are, and hence that we
must accept these as "given" in any process of deliberation.
How inconsistently such a position is usually held is evi-
dent from even a cursory examination of Aristotle's own
Nicomachean Ethics. In that work, Aristotle, instead of
directing his main effort to determining the means to happi-
ness, which he considers the universal end of human action,
expends his energy in trying to ascertain what the true
nature of happiness is. From a practical point of view and —
Aristotle's aim in the Ethics was thoroughly practical such —
an analysis would be entirely pointless were it the case that
the goal or end to which all men aspire is unambiguous, well-
formed, and fixed. The fact of the matter is that the work of
most great ethical thinkers has been directed not to ques-
tions of means but rather to the still more difficult task of
clarifying the primary aims of the good life. And their own
disagreements, uncertainties, and confusions are themselves
sufficient evidence that whatever else may be true of them,
our fundamental ways of life are neither clearly under-
stood nor universally aspired to. Professor C. I. Lewis puts
the point succinctly when he says, "At least half of the
world's avoidable troubles are created by those who do
not know what they want and pursue what would not satisfy
them if they had it." A similar thesis is implicit in the find-
ings of contemporary psychoanalytic and psychiatric theory.
If Freud and his successors are even half right, most human
action governed by hidden motives that are imbedded in

is
substitute symbols whose practical import we come to un-

derstand only with the greatest difficulty. Beyond all this,
however, the lives of even "normal" human beings are largely
controlled by words and symbols whose meanings are fre-
quently ambiguous or misunderstood. It is because of this
that the task of clarifying such golden words as "liberty,"
"justice," "democracy," "person," and love" is so essential to
the well-being of any people whose way of life is expressed
in terms of them. For if they are unclear or confused or in-
consistent, then so also is the way of life.
no part of my intention,
It is as my allusion to psycho-
analysis might seem to indicate, to suggest that philosophi-
cal analysis can provide a complete therapy for mankind's
ethical neuroses. What I do contend is that the clarification
of ideals and the clarification of the terms by means of which
they are articulated is a single process, and that in a humble
way, every attempt to determine what sort of life one really
wants to five involves essentially the same painful process of
analyzing and clarifying the meanings and uses of words to
which the analytical philosopher devotes himself profession-
ally. The difference between them is not so much a difference
in method as a difference in the thoroughness and sensitiv-
ity with which it is applied.
The point I am making might itself provide the topic for
another book. It must suffice here to conclude this discussion
with a few observations that may further illustrate my
theme. Consider, for example, how frequently the hesitancy
and confusion of our practice is correlated with lack of un-
derstanding of the import of ideals to which we think we
have committed ourselves. Consider how often practical
disagreements not only between individuals but also be-
tween whole societies are due to failures to recognize equivo-
cations in the terms by means of which conflicting aims are
expressed, and how baffled we frequently are as to the
causes of such disagreements. Practical misunderstanding
and conflict constantly arise from the fact that the same
terms are unwittingly used by both parties to express differ-

ent or opposing aims, or, perhaps more tragically, from the
fact that different terms are unknowingly used to express
what are would be common aspirations if only the terms
or
themselves were better understood. Communists and liberals
alike talk about something they call "democracy"; both ap-
pear to regard it as the consummation devoutly to be wished,
differing apparently only with respect to the means to be
employed; yet how profoundly unlikely such a characteriza-
tion is. Or again, and on another plane, such different moral-
ists as Epicurus, Marcus Aurelius, Jesus, Kant, Hobbes,
and John Stuart Mill expressed their respective conceptions

of the good life in terms which at least superficially express
different and in some cases opposing moral attitudes. How
can the man of pleasure, the stoic, the man of God and
Love, the incarnate rationalist, the incarnate egoist, and the
utilitarian all be supposed to aspire to the same ultimate
human condition? I do not say they can; I do maintain that
the only way such a question could be answered is through
the same sort of analysis which, on another level and in
another context, is employed in trying to determine the
meanings of "ought" or "good" and the logic of moral justi-
fication.
My contentionis, then, that both the methods and the
results of contemporary analytical ethics are indispensable

adjuncts of enlightened moral practice, and that their study
is therefore a useful or even a necessary part of any truly
humane or liberal education. What this also means, if I am

right, is that the self-conscious analytical philosopher has not
at all abandoned philosophy's ancient search for wisdom, but
on the contrary is contributing his own important share to
the world's all too skimpy fund of practical understanding.
He makes his contribution partly by providing us with
sharper tools and a clearer notion of the search itself, and
partly in a more direct way by freeing us from ancient
myths and fetishes which have created endless confusion and
needless disagreement about matters that are not necessary
parts of the tragedy of human no other way,

existence. If in
he would have done his share toward the advancement of
human enlightenment and freedom by showing us how and
on what terms morality can be a part of action. And his only
but mortal enemy is the obscurantist and the mystagogue.
[ii]
The Multiple Roles of the
Language of Conduct
FALSE ISSUES IN MORAL PHILOSOPHY
In an essay summarizing the outstanding issues in contempo-

rary moral philosophy, 1 Professor William Frankena divides
the ethical theories now prevalent into three main types:
(a) which
"naturalistic" theories assert that ethical state-
ments are cognitive and true or false and deny that such
statements are non-descriptive or that ethical terms desig-
nate any unique, non-natural characteristics; (b) "intui-
tionistic" theorieswhich agree with naturalism that ethical
judgments are cognitive and true or false, but hold that
ethical terms do designate at least one unique, non-natural
characteristic and that they are ( in the empirical sense ) non-
descriptive; and (c) "noncognitive" theories which agree
with intuitionism that ethical statements are non-descriptive
and with naturalism that ethical terms do not name any non-
natural characteristics, but deny that ethical statements are
cognitive and true or false. This is a very fair statement of
the essential contents of these theories, and a very neat
statement of the basic issues which divide — or appear to
divide —moral philosophers at the present time.
Under each main rubric there are of course many sub-
which shade by degrees
theories into one another. Under
1 William Frankena, "Moral Philosophy Mid-Century," The Philo-
at
sophical Review, Jan. 1951, pp. 44-55.
naturalism one finds the many pure and mixed types of

voluntarism, hedonism, self-realizationism, and pragmatism.
Under intuitionism one finds some philosophers holding that
there is only one non-natural characteristic, others that there
are more than one; some who define "right" and "ought" in
terms of "good," some who define "good" in terms of "ought,"
and others who hold that they are mutually indefinable.
Under "noncognitivism" one finds expressionists, individual-
istic imperativists, collectivistic imperativists, "rational" im-
perativists, ceremonialists, conventionalists, performists, and
so on. Indeed, we are only beginning to see how wide is the
variety of theories which fall under the general and mis-
leading heading of "noncognitivism."
But as one reflects upon Professor Frankena's classifica-
tion a number of disconcerting facts begin to emerge. 2
For
one thing it provides no place for the eclectics whose analy-
ses of moral discourse cut across party lines. Mixed theories,
like their ideological counterparts, rarely make the philo-
sophical headlines. It worth noting, however, that some of
is
the ablest thinkers in recent ethics have implied that no

one of the above theories does justice, in their opinion, to the
diversity of statements in moral discourse. For example,
there are intimations that while Professor Frankena himself
appears to be sympathetic with certain intuitionistic analyses
of "ought," he is inclined toward a naturalistic, perhaps even
a hedonistic analysis of "value." There is also the evidently
mixed view of C. I. Lewis, who distinguishes sharply be-
tween "valuations," which are regarded as empirically veri-
fiable statements of fact, and "ethical judgments," which,
one gathers, are to be construed as imperatives of a certain
sort. One recalls also that Professor Stevenson, whose earlier
papers upheld a relatively pure version of the expressive-
incitive theory, now holds that ethical terms have dual mean-
ings which are partly descriptive. Even A. C. Ewing, the
most intransigent advocate of intuitionism in the recent lit-
2 I do not doubt that Professor Frankena himself is well aware of

them. His classification was intended for convenient mapping purposes
only.
The Multiple Roles of the Language of Conduct 35
erature, agrees that many senses of such terms as good" and

"ought" are empirically descriptive. An excellent case could
be made, moreover, for the thesis that such older intuition-
ists as Moore and Prichard held mixed theories. Their in-
tuitionism seems to have been directed exclusively to some

particular sense of "good" or "ought" which they considered
"characteristically ethical."
Reflecting further, one finds that Frankena's classifications
are misleading. For they may suggest that writers who are
formally affiliated with a "naturalistic" account of ethical or
value judgments really deny that such judgments have "non-
cognitive" effects or functions —
or, if one prefers, "meanings."
Such is far Thus, long before the vogue of the
from true.
emotive theory, R. B. Perry, the arch-defender of the empiri-
cal meaningfulness of "value" and the empirical verifiability
of value judgments, attempted to account for the practical
relevance and "moving appeal" of such judgments. 3 By this
he meant explicitly their capacity to incite interest. And
John Dewey, who also defends the "scientific" status of
valuations, also distinguishes them as de -jure, as distinct from
merely de facto judgments. Whether any of the naturalists
has successfully accounted for the non-descriptive or "nor-
mative" aspects of moral discourse is, to be sure, extremely
debatable. But it is plain that most of them did not mean to
deny that it has such aspects. On the contrary.
It should be borne in mind that until the last two decades,
it was not the practice among philosophers of any school to
regard the "noncognitive" intentions or effects of discourse
as part of its "meaning." When such philosophers attempted
to analyze the "meanings" of ethical terms they automatically
concerned themselves with their purported significations
alone. Even the earlier versions of the emotive theory itself
tended to regard questions concerning the "meaning" of
ethical judgments exclusively in descriptive terms. Hence
3 R. B. Perry, "Value and its Moving Appeal," The Philosophical
Review, XLI, 4 (July 1932), pp. 337-50. Perry's awareness of the emo-
tive meanings of such words as "good" is quite evident, both in this
article and in his "Value as Simply Value," The Journal of Philosophy,
XXVIII, No. 10 (Sept. 1931), pp. 519-26.
the paradoxical thesis of Carnap and Ayer that ethical judg-

ments are "meaningless." When we probe beneath the sur-
face, however, we find that so staunch an emotivist as Ayer
admitted from the outset that such words as "good" some-
times are used "cognitively." 4 I do not suggest, of course,
that Ayer and Perry or Dewey are in complete agreement. I
do maintain that, when full account is taken of their views
concerning the functions and effects as well as the "mean-
ings" of ethical terms and judgments, it is misleading to
classify the former exclusively as an emotivist or the latter
exclusively as naturalists.
Pondering the literature of moral philosophy during the
past half-century one is forced more and more to the con-
clusion that although there may be some real issues, many of
the issues that divide philosophers are really pseudo-issues.
By this I mean that they often arise from concealed and mis-
leading terminological disagreements at the meta-ethical
level. Such meta-ethical expressions as "ethical term," "nor-
mative," "characteristically ethical," and so on, are not ex-
pressions having well-established or unambiguous meanings
in ordinary language. If they are to be significantly em-
ployed, therefore, they must be stipulatively defined, either
by enumerating the particular words or expressions in or-
dinary language to which they apply or else through some
descriptive phrase which clearly specifies some particular
class of expressions without regard to the individual words
which fall under it. Butthis is just what normally is not done.
One is left to infer as best one can from the context what
this philosopher means by "ethical term," or that philosopher
understands by "normative" or "characteristically ethical."
And when one attempts to determine from their examples or
from the which their respective analyses
sorts of "criteria"
are designed to meet, becomes evident that the terms, or
it
at least the meanings of the terms with which they are con-
cerned are not always the same. One philosopher is pre-
occupied with the problem of accounting for die rhetorical
4 A. Ayer, Language, Truth and Logic (London: Gollancz Ltd.,
J.
1936), pp. 102-3.
or incitive aspects of statements in which such terms as

"good" and "ought" occur, another with their informative
aspects, still another with their critical or educational roles.
Again, one philosopher is concerned with the "meaning" of
"ought" in such a statement as "You ought not to have made
a promise you didn't intend to keep"; another with the an-
alysis of the whole class of statements, whatever the terms
in which they may be formulated, that express attitudes; and
yet another with the innocuous thesis that statements about
desires and enjoyments and their necessary means are
empirically verifiable.
The upshot of such reflections is that most of the sup-
posedly opposing theories of ethics are but limited accounts
of different "moments" of a complex and variable mode of
discourse which serves, and indeed must serve, a wide
variety of functions. As one reads a competent defense of any
of the main theories one cannot help acknowledging that
here is a kernel of truth, misleadingly stated. It is only when
it is taken to imply a denial of what other theories assert that
it appears indefensible. The various "criteria" which have
been proposed for an adequate analysis of moral discourse

turn out, on such an interpretation, to be nothing more than
expressions of the different perspectives from which such
discourse may be viewed. From a certain point of view each
criterion is legitimate, and the attempt to provide analyses
capable of meeting it often yields interesting and important
accounts of certain otherwise neglected aspects of moral dis-
course. Yet none encompasses all of the roles which such
discourse is called upon to fulfill in ordinary life. It would
seem to follow, therefore, that the better part of phil-
osophical good sense in moral philosophy would be to divide
the spoils among the several theories, and to take seriously
rather than merely to acknowledge formally the fact that in
this sphere above all others our language is extremely flex-
ible in function, purpose, and effect. In this light, I propose
an end to, or at least a moratorium upon, futile disputation
with imaginary opponents, and that we cease pitting part
against part or the special perspective against the total con-
text. Why should we spend our strength in senseless polem-

ics against "types" of ethical theory, which at best are
special theories of particular aspects of the language of con-
duct, when, without sacrificing any theoretical position worth
defending, we may divide and conquer at our leisure?
THE COMPLEXITIES OF PRACTICAL DISCOURSE

Leaving this question to echo and, I hope, re-echo in the
mind of the reader, let us turn from type theories of "good"
and "ought" to the actual complexities of practical discourse
itself.Here one discovers that whatever may be the truth
about so-called "ethical terms" any adequate language of
conduct must in principle provide a whole battery of con-
cepts, rules, techniques, procedures, and other linguistic
functions. To accomplish the purposes both of individual and
of social life it is, at the very least, necessary that we be able
to do the following things with words: (1) identify objects,
(2) describe their characteristics, (3) predict their causes
and effects, (4) infer from such descriptions and predictions
what means are necessary to accomplish our ends and what
ends are worth entertaining seriously, (5) express individual
and collective attitudes and decisions, (6) adjudicate dif-
ferences in attitude, (7) clarify aims, (8) commend, exhort,
and persuade others to acts which we approve or deem de-
sirable, (9) make
promises, (10) assign responsibilities, (11)
authorize, regularize, and correct behavior, and this on a
variety of levels. If these requirements are to be met, we
would then have to be able to formulate definitions, state-
ments of fact ( of various sorts ) , counter-factuals, arguments,
rules, commands, and principles. In short, an
proposals,
adequate language of conduct would have to be able to
marshal and deploy the entire gamut of human capacities.
And if such a language did not exist, it would have to be in-
vented.
The only question is whether ordinary language success-
fully mirrors the complexities of man's practical life. And as
one turns back now and reflects upon the great variety of
ethical theories and the great number of "senses" of ethical

terms and roles or functions of practically oriented judgments
which such answer is
theories reflect, the only intelligible
that ordinary language provides at any rate a vastly richer
and more supple instrument for expediting the conduct of
life than any ethical theory represents it to be. At certain
points, no doubt, this language is ambiguous; at others it

both permits and requires contextual qualification in order
to function properly as a vehicle of appraisal. There is no
more reason why ordinary language should be worshiped
as a sacred cow in the sphere of valuation than in that of
exact science. But, before attempting any philosophical re-
construction of it, we should see what this language is al-
ready capable of expressing. If we do we will find that in
broad outlines it fairly represents the multiplicity of func-
tions which, in the sphere of human conduct, are required
of it.
I we adopt a more flexible, and plu-

propose, then, that
ralisticapproach to the study of the language of conduct. As
here conceived, such a study would be prepared not merely
to acknowledge but also to explicate and relate the whole
gamut of "meaning" and the whole ensemble of roles to be
found in this form of discourse. One set of meanings would
then be considered as "essential" as any other; none would be
regarded as more "characteristically" ethical than the rest.
On such a view, "system" would be held suspect, if by this is
meant the attempt to formulate sets of definitions through
which all of the meanings of ethical terms are analytically
reduced to some supposedly "basic" concept of value. Nor
would the varieties of practical judgment and argument be
assumed to be merely species of a common genus. The
planting of axiological trees of Porphyry, to judge from the
literature, is not a fruitful occupation. At the very least
there would have to be an orchard, and a good many graft-
ings of branches before we could begin to cope with our
problems. What is wanted is the successive elucidation of
distinctbut practically related levels or dimensions of moral
discourse, and an explication of the movement of evaluation
from reaction to gesture, from expression to demand, from

puzzlement to deliberation and choice, from preference to
criticism, from the envisagement of means and the pre-
diction of consequences to the clarification or definition of
ends.
It is not enough, however, to treat this movement as a
series of separate "contexts" within which some particular
meaning or role is elicited. Moral discourse, being practical,
is free to draw upon whatever intellectual and motivational
resources are necessary to clarify, appraise, alter, and or-
ganize its ends-in-view. Its ends-in-view exclude no interest
or purpose to which human flesh is heir. It is the failure to
acknowledge explicitly the dialectical fluidity and movement
of practical discourse that is in part responsible for the sense
of unreality we experience in contemplating the formalized
what we are supposed to be
philosophical reconstructions of
doing when we engage in moral deliberation and criticism.
The "examples" of moral discourse on this or that level which
abound works on ethics resemble nothing so much as the
in
elegant but artificial period rooms which we see displayed
in museums of art. In such a room each piece is a perfect

example of its style; every appointment a marvel of con-
gruity and fitness. Nothing is permitted to be "out of con-
text." An actual living room, however, is a very different
sort of thing. Its harmony is "functional," and its functions
determined only in part by foresight and purpose. To the
outsider, it often appears disorderly, confused, eclectic, per-
haps even in bad taste. Nevertheless, to the person who uses
and lives in it, its parts fit together in a way which is prac-
tically intelligible even if, taken as a whole, lacking in delib-
erate design. Similarly, actual moral arguments frequently
involve a series of statements of different logical types which
are nonetheless related to each other in ways that are prac-
tically relevant to the successive and shiftingdemands of the
participants.
This is not intended to imply that there is no "form" to
moral discourse, or that the philosopher, like the misguided
critic who tries to recreate the aura of a work of art, should

attempt to simulate the "infinite" flow and variety of moral
discourse. On the contrary. As I have tried to show in other
essays included in this volume, there are, on at least some
moral discourse, governing principles of relevance
levels of
by means of which we can contain and legitimately correct
an argument which tends to get unmanageable. In morals as
well as in logic or science there are distinctions between ap-
pearance and between inference and random as-
reality,
sociation, and between and persuasion. But prac-
justification
tical argumentation as a whole has to satisfy interest, not a
book of rules. By necessity, therefore, it cannot be wholly
confined by the rules and procedures which may be in-
voked at particular stages to channelize and order it. It can-
not be too often insisted that the end product the ulterior —
—
decision is not clearly envisaged at the outset. A moral
choice, like a finished work of artistic creation, emerges by
degrees from a process of cumulative acts of expression,
clarification, prediction, justification, and persuasion. And
just because the real conclusion of a moral argument is a
practical decision, there is no end to the questions and an-
swers which at any point may become relevant to it. It
would be foolish to deny, as some writers have done, that in
deciding what we are to do, considerations of logical co-
herence or of empirical verifiability have their place. It
would likewise be foolish to deny that on certain levels of
deliberation, there are distinctive rules of relevance which
determine the order of "moral" justification. We have stand-
ards which enable us to tell a "good" reason in ethics, just
as we have standards which enable us to distinguish a valid
inference in logic. But just because they involve the applica-
tion of rules, these levels have limits which purposive activ-
ity as a whole does not. There just is no end to the possible
emotional, cognitive, and volitional considerations which
may have bearing upon our practical deliberations and
choices.
Here it is not a matter of what the rules of a supposed
"moral game" permit but of what the active human being is

prepared to accept, what his total fund of sensibilities and
capacities permit or require. In action, every consideration
is and must be subordinated to the exigencies of need. If
one cannot find what one wants by way of an answer on one

level of discourse, one must perforce search elsewhere. If
one's practical hesitations cannot be removed by arguments
which observe the amenities of one level, one is driven, willy-
nilly, to seek answers on another. The steps, logical or other-
wise, by means of which a critical conflict of interest is re-

solved are necessarily secondary to the controlling interests of
both parties in reaching a basis of agreement. When our prob-
lem is to discover a goal which we may hold worthy of our
effort, it is less important by what means
—
"ethical" or "un-
—
ethical" we arrive at our solution than that when we find it,
it be genuinely satisfactory to us. Rationality, whether in the
sphere of formal deduction, of factual prediction, or of

ethical criticism, has proved its utility to countless genera-
tions of men. We violate its canons at our peril. But there
are also occasions when free association may provide a
better method of resolving a practical difficulty than close
reasoning. When the problem is ennui or failure of nerve,
the piling up of "reasons" "justifications" may be ir-
and
relevant and even silly. What
wanted here is stimulation
is
and example, and this of the most moving sort. And since
rhetoric without imaginative objectification usually ends by
being counterpersuasive, poetry in such a case may be the
only effective means of moral persuasion. To acknowledge
that the reading of a novel or the witnessing of a play may
sometimes be a better way of reaching a decision than the
endless and paralyzing reappraisal of consequences is neither
to debase art nor to defend irrationalism in ethics; it is sim-
ply to recognize that the symbolic means by which we de-
cide what is be done are many, and that "nonrational"
to
procedures also have their proper function in the moral life.
Where reason may be invoked it may be "wrong" to ignore it,
but where there is no motive to abide by the rules of "right
reason," reasoning itself becomes pointless. In the last analy-

sis, any method must procedure to
justify itself as practical
interest. Failing this it remains morally quite without rele-
vance. For in the end, the "logic" of any practical discourse is
grounded only in human need.

—
[III]
Definitions, Factual Premises,
and Moral Conclusions
TWO FORMS OF THE NATURALISTIC FALLACY
"We have seen that the claim to infer significant ethical

propositions from definitions of ethical terms, which appears
to constitute the essence of what Professor Moore calls the
naturalistic fallacy, is a special case of a more general fal-
lacious claim, namely, the claim to deduce ethical propo-

sitionsfrom ones which are admitted to be non-ethical." x In
these words Mr. A. N. Prior sums up I think very fairly —
a view which many philosophers have come to regard as the
fundamental point of departure for any adequate analysis of
moral discourse. It, or something like it, is also popular
among social scientists who like to insist upon the ethical
"neutrality" of their disciplines. 2 Despite the fact, however,
that some of the best minds in recent philosophy have, for a
variety of reasons, maintained the view, it seems to me to be
correct only in certain very limited respects. And even in
these respects it is, as usually formulated, seriously mis-
leading. There are other, more important senses in which, as
1 A. N. Prior, Logic and the Basis of Ethics (London: Oxford Univer-

sity Press, 1949), p. 95.
2 For example, see Max Weber, The Methodology of the Social Sci-
ences, tr. by Edward Shils (Glenville, 111.: The Free Press, 1949), pp.
1-113. F° r nry part, Weber's prolonged discussion does little to clarify
the problems at issue. Weber's powers as methodologist seem to me
generally overrated.
—
Definitions, Factual Premises, and Moral Conclusions 45
I shall presently try to show, the contrary position must be

upheld. 3
Let us begin by reviewing those respects in which the
—
view I will henceforth refer to it as antidescriptivism
might plausibly be maintained.
Consider first the more general version, namely the thesis
that the claim to deduce ethical propositions from ones ad-
mitted to be nonethical is fallacious. Now by "ethical propo-
sition" let us here understand "any judgment containing such
words as 'ought,' 'right,' and 'desirable' which is used norma-
tively to prescribe that something is to be done or ap-
proved." And by "nonethical proposition" let us understand
"any statement of fact or of logic to which the predicates
'true,' 'false,' or 'probable' may be ascribed, and in which no
word functions essentially in the prescriptive or imperative

mode." Let us agree also that the laws of logic, by which
alone one statement can be validly deduced from an-
other, apply only to propositions which are true or false. 4
Granted these assumptions, it is not hard to see in what
sense it would be possible to say that ethical propositions
are not deducible from nonethical premises. And anyone who
maintained the contrary would seem to be involved in a
fundamental confusion of logical categories. 5
Consider next the narrower version of antidescriptivism,
namely the thesis that significant ethical propositions can-
not be validly inferred from definitions of ethical terms. For
the moment let us understand roughly by "definition" "any
statement in which the purely descriptive or logical mean-
3My reasons for this contention do not involve a defense of ethical

naturalism.
4 I am, of course, here using the terms "true" and "false" in their
stricter or cognitive sense. Other looser and more figurative senses are
irrelevant to the present point. In another place I will have something
to say about these senses as employed in moral discourse.
5 Careful reading of such versions of the emotive theory as that of
Professor Ayer suggests that perhaps nothing more was implied by the
proponents of the theory than I have stated in this and the following
paragraphs. The not entirely unjustified opposition to the view was
largely due to misleading formulations. See A. J. Ayer, Language,
Truth and Logic (London: Gollancz, 1936), pp. 105-6.
ing of an expression is stipulated, analyzed or otherwise

characterized." Here again, given the previous interpreta-
tion of "ethical propositions," it would be simply an egre-
gious blunder to try to infer a significant ethical proposi-
tion from the definition of an ethical term —assuming, for
the nonce, that it has one. Despite evident differences, the
blunder would be comparable in some ways to the error of
trying to infer a nonlinguistic statement of fact from the
mere analysis of a concept.
So far, then, the contentions of antidescriptivism, in both
its wider and its narrower versions, may be granted. To this
extent, we may agree that at bottom there really are basic

categorial mistakes inherent in the "naturalistic fallacy,"
even though the arguments intended to show this by Moore
and his followers are quite inconclusive. 6 Unless these state-
ments are immediately qualified, however, they tend to be
misleading. For they appear to convey the impression, and
indeed have conveyed the impression to a generation of
moral philosophers, that statements of fact provide no
logical support for ethical propositions, and that in no sense
can significant ethical propositions be legitimately inferred
from definitions of ethical terms.
SOME ROLES OF DEFINITION IN MORAL PHILOSOPHY
Let us see how such an impression has arisen. We may begin

by considering a number of the most influential conceptions
of "definition" in moral philosophy. First let us notice the
view expressed by Moore in Principia Ethica: "What, then,
is good? How is good to be defined? Now, it may be thought
that this is a verbal question. A definition does mean the ex-

pressing of one word's meaning in other words. But this is
not the sort of definition I am asking for. . . . My business

is solelv with that object or idea, which I hold, rightly or
wrongly, that the word is generally used to stand for. What I
6 See William Frankena, "The Naturalistic Fallacy," Mind, 1939,

pp. 472 See also A. N. Prior, op. cit., pp. 1-12, 95-107.
ff.
." 7
want to discover is the nature of that object or idea. . .
And was asking for,

again, "definitions of the kind that I
definitionswhich describe the real nature of the object or
notion denoted by a word, and which do not merely tell us
what the word is used to mean, are only possible when the
object or notion in question something complex." 8 A
is
similar conception is to be found in R. B. Perry's General

Theory of Value: "No one would be disposed to deny that
there is a common something in truth, goodness, legality,
wealth, beauty and piety that distinguishes them from gravi-
tation and chemical affinity. It is the express business of
theory of value to discover what this something is; to de-
fine the genus, and discover the differentiae of the species.
By means of such definitions and systematic connections
theory of value may unify the special philosophical and so-
cial sciences enumerated above and arbitrate between
them." 9 Whether either of these conceptions of definition is
defensible as it stands we fortunately do not now have to
consider. What is of interest to us here is that both of them
regard definition as a purely explicative operation which is
to be performed only on the descriptive or referential mean-
ings of words. Other functions of language and other forms
of clarification are simply ignored.
We are now faced with an interesting dilemma: if, on the
one hand, we insist on construing "definition" as Moore
and Perry do, many statements occurring in moral philos-
ophy that are normally taken as definitions simply are not
really so at all; but if, on the other hand, we construe such
statements as bona fide definitions, then these theories of
definition must be regarded as faulty, since they completely
mischaracterize the statements in question. In either case,
however, we must conclude that the intentions of many
moral philosophers who have made such statements have
been radically misunderstood. And the source of the mis-
7 G. E. Moore, Principia Ethica (London: Cambridge University
Press, 1929), p. 6.
8 Ibid., p. 7. Cf also
.
pp. 8-9.
9 R. B. Perry, General Theory (New Longmans,
of Value York:
Green, 1926), pp. 4-5.
understanding must be attributed to those who, like Moore,

have carelessly applied the term "definition" to expressions
whose roles are not to explicate the descriptive meanings
of words. It is these analysts who are responsible for the mis-
leading implications of the thesis that the claim to infer an
ethical conclusion from the definition of an ethical term is fal-
lacious. For although, as we have seen, they are correct in
maintaining that in Moore's sense of the term "definition"
ethical conclusions cannot be inferred from definitions of
ethical terms, they mislead us into supposing that those
charged with commission of the fallacy actually did claim to
draw such inferences. The truth is that most of the moral
philosophers charged with making this claim were really do-
ing something utterly different. To be sure such philosophers
often did purport to draw ethical conclusions from "defini-
tions" of ethical terms (together with other premises which
were usually factual). But their definitions were not, and
were never intended to be construed as descriptive defini-
tions in any of the senses considered above.
For purposes of illustration, let us briefly consider the
ethical theory known traditionally as hedonism. I shall dis-
cuss it in the form popularized by Mill and subsequently re-
formulated and criticized by F. L. Bradley, Moore, and their
followers. By hedonism let us understand here the doctrine
which holds that "happiness (or pleasure) alone is good" or
that "pleasure is desirable, and is the only thing that is de-
sirable as an end." I may say, in passing, that I have no in-
terest here in any of the reasons that might be given, or in
fact were given by Mill and other writers, in defense of this
doctrine. Nor do I wish to suggest that the doctrine is de-
fensible. At this point, my concern is merely with the al-
legation that the hedonist is wrong when he infers from the
thesis of hedonism that in particular situations we ought al-
ways to choose that alternative which will produce the most
happiness or the least suffering. The thesis itself may be
vulnerable; that is beside the point. What I maintain is that
no fallacy whatever is involved in making the inference.
In order to show this, let us see how Mill himself regards
the thesis. First of all, it is fairly clear even in Utilitarianism

that he does not regard it as a theoretical statement of any
sort, whether analytic or synthetic. As he says repeatedly
in this work, 1 it is a "principle" or "rule" of conduct. Morality,
like legislation, is a "practical art"; its rules, as distinguished
from the laws of science, are "practical principles" for the
conduct of life. 2 This point is elaborated with the greatest
clarity in the Logic, a work which, in this connection, none of
his critics appears to have noticed. In that work, it may be
remembered, Book VI is "The Logic of the Moral Sci-
ences," at the end of which Mill devotes a chapter to what he
calls "The Logic of Practice or Art, Including Morality or
Policy." 3 In this chapter, he begins by sharply distinguish-
ing "moral science" —we would now call it "social science"
—which concerned with inquiries into "the course of na-
is
ture," from that "inquiry the results of which do not express

themselves in the indicative, but in the imperative mode, or
in paraphrases equivalent to it; what is called the knowledge
of duties, practical ethics, or morality." 4 So understood, he
tells us, the rules of morality do not consist in assertions
respecting matters of fact. "The Method, therefore, of ethics,

can be no other than that of the Art of Practice." 5 Now it is
the function of Art to propose an end. This end having been
proposed, it is handed to the sciences, which in turn treat it
as an effect. They explore its causes, and then send it back to
Art with a theorem which states the circumstances and
conditions under which it could be produced. Art in turn
considers these and asserts on its own authority that the at-
tainment of the end is desirable. "But though the reasonings
which connect the end or purpose of every art with its
means, belong to the domain of Science, the definition of the
1 S. Mill, Utilitarianism, Everymans Library ed. (London: M.

J. J.
Dent & Sons, 1944), pp. 2, 11, 32, etc.
2 Ibid., Compare also the crucial first paragraph of the much
p. 2.
criticized fourth chapter, p. 32.
3
J. S.
Mill, A
System of Logic, 8th ed. (London: Parker, Son, &
Bourn, 1862), Vol. pp. 652-9.
II,
4 Ibid., 653. (Italics mine.)
p.
5 Ibid.
)
end itself belongs exclusively to Art, and forms its peculiar

province." 6
And then, as clearly as can be, "Propositions of
science assert a matter of fact: an existence, a co-existence, a
succession, or a resemblance. The propositions now spoken of
[those of art] do not assert that anythingbut enjoin or is,
recommend that something should be. They are a class by

themselves. A proposition of which the predicate is ex-
pressed by the words ought or should be, is generically dif-
ferent from one which is expressed by is or will be." 7
Mo-
rality, Mill maintains, is the Art of Life, and its supreme
principle, of course, is for him the thesis of hedonism. This
thesis defines the end of conduct, and provides the basis for
justifications of "secondary" moral rules that prescribe our
particular duties, and through them for the justification of
ethical propositions that prescribe what we should do in par-
ticular cases.
So far, then, is Mill from committing the real fallacies in
question, that he himself formulates, with the greatest clarity
and elegance what they might be presumed
of language, just
to consist in. To be
he claims to infer significant ethical
sure,
conclusions from that definition which identifies the end of
the Art of Life; but this is because the definition itself
"enunciates the object aimed at, and affirms it to be a de-
8
sirable object." The definition does not provide a logical
analysis of "desirable" or "good" — at least as "logical analysis"
is now understood. This analysis is provided in his general
account of "The Logic of Practice." The definition itself is
nothing but the statement or, better, the enunciation of a

norm or standard of conduct. And the "inferring" of signif-
icant conclusions from it is the business not of theoretical
but of practical reason. 9 A fallacy would be involved here
only if one attempted to infer an ethical conclusion from defi-
nitions concerned essentially with the explication of the
descriptive meanings of symbols.
6 Ibid., p. 656. ( Italics mine.

7 Ibid.,
pp. 656-7. (Italics in text.)
8 Ibid., p. 656.
9 Cf . ibid., p. 657.
If we return now to Utilitarianism and interpret it, as Mill

appears to have intended, not as a meta-ethical theory or
analysis of moral discourse but as an enunciation of his
x
doctrine of hedonism and as a defense of it against practical
objections,we will find that many (although not all) of the
supposedly questionable or fallacious "claims" of that work
involve no serious theoretical confusions whatever. On the
point at issue, he is, I believe, quite consistent. To say, there-
fore, that for Mill the principle of hedonism constitutes a
definition of the desirable, is to say in effect that it is ipso .
facto and by intention a practical or, if you like, a "per- ]
suasive" definition. But there is no confusion. Nor is there '
any deception. The imputation of confusion or deception

arises solely from the mistaken and misleading application of
descriptive definitional models in the interpretation of what
plainly and explicitly is a normative ethical system. Given
his own definition of desirable he is perfectly justified in in-
ferring from it, in conjunction with the relevant "theorems"
of science, those conclusions which prescribe in particular
just what we ought to do. The categorial blunder, I submit,
is not but that of his critics.
Mill's,
Now can imagine that certain readers may conclude
I
from all this that we should have seized the first horn of
our dilemma in the first place, namely, that many state-
ments occurring in works of moral philosophy that are nor-
mally interpreted as definitions simply are not so at all. I do
not agree. Acceptance of such an alternative rests upon an
essentialist attitude toward definitions which I reject. But
even if for the sake of argument we accept it, it still does not
follow that any confusion is involved in Mill's doctrine on
the point in question. For one cannot charge someone else
with the commission of a fallacy on the ground that he
claims to infer an ethical conclusion from a definition of
ethical terms, when it is as plain as can be that his so-called
definition was designed for a purpose very different from
that which animates the conception of definition implicit
in the first horn of our dilemma. Even more to the point,
1 Cf. Utilitarianism, p. 5.
one had no right to suppose that, by means of such a con-

ception of definition together with ordinary rules of logic,
one can shed any on the process involved in inferring
light
ethical conclusions from what are normally interpreted as
definitions by moral philosophers. Indeed, by so construing
definition one thereby puts well-nigh insuperable diffi-
culties in the way of properly interpreting what moral
philosophers say and mean. By what is perhaps nothing
more than a process of association of ideas one is led,
when one sees the word "definition," or when one sees some-
thing that looks like a "definition," to an interpretation of it
which at once utterly misses the point and results in a mis-
taken charge either of logical confusion or else of down-
right intention to deceive.
A more profitable procedure is to resist the temptation
to cut the Gordian knots of meaning by essentialist definitions
(they might as well be called stipulations) and to recognize
instead the ambiguities and vagueness inherent in the normal
use of "definition" and the variable functions of actual defi-
nitions. In this way one approach a certain type of dis-
will
course without preconceptions concerning the model to
which it ought to conform. And perhaps, if one is lucky, and
manages to keep all the necessary distinctions in mind, one
may be able to perform the difficult philosophical task of
correctly interpreting a form of discourse. What is wanted
is not an immediate cry of confusion, but the illumination of
a normal function of language; not the implicit attribution of

bad faith or of obscurantism, but the explanation of the na-
ture of symbolic processes that answer to other needs than
those of science or formal logic.
Any normal reader of Plato, Epicurus, or Mill recognizes
at once that, for the most part, he is reading in moral philos-
ophy, not about it. Such a reader, I submit, takes their
"definitions" and "inferences" for what they are, namely, prac-
tical definitions of norms and practical reasonings to par-
ticularmoral conclusions. I agree with Professor Stevenson
and Professor Robinson that the Republic is a prolonged es-
2
say in the "persuasive" or practical definition of "justice."
For better or worse, Plato was desperately concerned to
clarify and to defend a system of norms upon the acceptance
of which, as he believed, the salvation of Atiiens depended.
But there was nothing deceptive or illicit in what he was
doing. The term "justice" no doubt has a strong aura of
"emotive meaning." And like anyone else who attempts to
define a norm in terms of "justice," he inevitably "traded"
on its emotive meaning. But that he was proposing an "ideal,"
not explicating the use of a term in his famous definition
of justice, seems to me as clear as day.
Therefore, I do not at all agree with Professor Robin-
son's remark, "A persuasive definition, it may be urged, is at
best a mistake and at worst a lie, because it consists in get-
ting someone to alter his valuations under the false im-
pression that he is not altering his valuations but correcting
his knowledge of facts. I am tentatively inclined to accept
this view,with the practical conclusion that we should not
3
use persuasive definitions." I do not say that it never hap-
pens that emotive meanings are used illicitly to modify at-

titudes. Of course it does. I do say that in the context of
normative ethics, where understood that a writer is
it is
making practical proposals for the guidance of conduct, no

confusion and no deception is involved.
This is why I should prefer, in discussing Plato or Mill, to
use the phrase "practical definition" rather than "persuasive
definition" to characterize what they are doing. Perhaps un-
fortunately, the latter expression appears to suggest a process
which involves confusions and lies. And it does so, it should
2 C. L. Stevenson, Ethics and Language (New Haven: Yale Univer-
sity Press, 1944), pp. 224-6; Richard Robinson, Definition (London:
Oxford University Press, 1950), p. 166.
3 Robinson, op. cit., p. 170. It is instructive, I think, to see that in
so many words Professor Robinson has drawn a "practical conclusion"

from what he believes to be the statement of a fact. I applaud the
inference, and, given his belief about the effects of persuasive definition,
I consider it to have been his duty to have drawn it. If I do not make
the inference, it is because I take a different view of the definitions in
question.
be emphasized, precisely because of the tendency of analysts

to approach moral philosophers with descriptive models in
mind as the norms for correct definition and valid inference.
As I understand it, a practical definition is explicitly and
openly intended as the statement of a norm. Since it is
stated in ethical terms it also may be emotively loaded.
But this cannot be avoided in any case. No harm is done so
long as we know what is happening.
ON MISLEADING CLASSIFICATIONS
OF ETHICAL THEORIES
I wish, now, to say a word about current classifications of
"ethical theories." In my judgment most of them are mislead-
ing; and they are so precisely because they appear to be set
up for the purpose of classifying theories concerning the na-
and moral judgments, whereas the doc-
ture of ethical terms
from historical works in
trines so classified are abstracted
moral philosophy, the purpose of which was to define the
right or the good only in the normative or practical sense.
Thus, for example, Professor Broad classifies "ethical theo-
ries" in accordance with certain principles that distinguish
among them according to their conceptions of "ethical charac-
teristics." Clearly his classification concerns only metaethical
theories, not substantive normative systems. But then he ap-
plies his classification to such a writer as Mill who, in the
only work in moral philosophy to which his critics ever
refer, was doing ethics, not talking about it, and whose
definitions were definitions of ends, not theoretical analyses
of ethical "characteristics." Broad, in accordance with the
standard view, says that "Mill presumably meant to be a
naturalistic hedonist. But it is difficult to be sure in the case
of such an extremely confused writer that he really was
one." 4
And yet, if Logic or at those passages
one looks at the
in Utilitarianism in which Mill does talk briefly about moral
judgments, it is, unless I am badly mistaken, quite evident
4 C. D. Broad, Five Types of Ethical Theory ( New York, Harcourt,
Brace, 1930), pp. 258-9.
that, in Broad's sense of the terms, Mill is neither a naturalist

nor a hedonist in his theoretical ethics. Rather, he was pri-
marily a noncognitivist with a strong prejudice against de-
scriptivism or "naturalism." His affiliation is with the emotiv-
ists and imperativists, rather than the naturalists. Only in his
normative ethics is Mill a hedonist of sorts.
Indeed I am not at all sure that one can find a clear case
of a naturalistic hedonist in Broad's sense, i.e., one who be-
lieves that "good" is analyzable without remainder in terms
of "pleasure." A number of eighteenth-century writers do
seem to have held that "good" in its "natural" use does
mean the same thing as "agreeable" or "pleasant." But they
were usually from "moral
at pains to distinguish this use
goodness" which, for them, carries with it normative or prac-
tical implications. Even Bentham, who said in his downright
way that the words "ought" and "right" have a meaning
only when applied to actions conformable to the principle of
utility, is only an apparent exception. 5 As Mr. Stuart Hamp-
shire has suggested, it was really Bentham's concern to re-
place all this fiddle-faddle of "morality" with scientific so-
cial engineering guided exclusively by the social norm of
greatest happiness. 6
But perhaps even these remarks may be indirectly mis-
leading. For in saying that the great classical moral philos-
ophers were not primarily interested in "the analysis of
ethical terms" I may have created the impression, which I
most certainly wish to avoid, that the term "analysis" should
be pre-empted in ethics for the kind of tiling that analytical
ethicists generally do nowadays. 7 I believe that the effect
would be altogether desirable if philosophers somewhat en-
larged their conceptions of the roles of analysis itself so as to
acknowledge normative and practical functions.

explicitly its
In my judgment, the clarification of aims is at least as im-

portant a task for analytical philosophers as is the descrip-
See British Moralists, ed. by L. A. Selby-Bigge (Oxford: The
5
Clarendon Press, 1897), Vol. I, p. 342.

6 Stuart Hampshire, "Fallacies in Moral Philosophy," Mind, 1949,
P- 473-
7 Professor Karl Popper is a notable exception.
—
tion and elucidation of moral discourse itself. However

much one may disagree with it in detail, Professor Popper's
The Open Society and Its Enemies8 is a brilliant example
of what the philosopher may do in the way of clarification of
purposes and in sweeping away the manifold confusions
which produce practical misunderstanding.
Disagreement at the practical or moral level frequently
arises, not so much from differences of opinion about mat-
ters of fact or from well-defined and outright conflicts in
interest, as from confusions of language and thought and
from ambiguity and vagueness in the formulation of aims,
and hence in the aims themselves. Hedonists, for example,
have been called "pig philosophers" by those who, had they
only understood them, might have found more with which
to agree than to disagree. And on the other hand, as we shall
see, hedonists have frequently provoked disagreement by
faulty statements of their fundamental principle. The in-
terminable disagreements in contemporary moral philosophy
among those who have called themselves "hedonists," "volun-
tarists," "eudaemonists," and "instrumentalists" are often due
not to well-defined differences of aim as to opposing expres-

sions of what, if we could get clear about them, might turn
out to be common aims.
The reason for this is due largely to the vagueness and
the ambiguities latent in such notions as desire, will, satis-
faction, pleasure, pain, end, feeling, attitude, and belief

precisely the terms, in short, in which basic norms have so
often been expressed. It is, or should be, the business of
the moral philosopher not merely to codify and adumbrate
the principles by which we are to live, but also to remove
misconceptions standing in the way of their acceptance, and
to provide them with such determinateness of meaning as
may enable them to function adequately as practical guides
to conduct.
Only by such processes of reflection or analysis are we at
last able to grasp what we are enjoined to accept as right or
8 Two volumes (London: Routledge and Kegan Paul, Ltd.), 4th
edition.
good; only so can we intelligently deliberate on the ap-

propriate means to its realization. It is simply not true that
the only function of deliberations or practical reason is to de-
vise means to the fulfillment of already given ends-in-view.
As often as not it is the means that are clear, not the end.
Not knowing what is wanted, or confusion as to its nature, is
at least as common as, and perhaps an even more tragic
privation than not knowing how to acquire a known goal.
Utilitarianism, as it stands, is unquestionably inadequate as
a moral philosophy. But, comparatively speaking, its failure
is due mainly fundamental lack of clarity with
to a very
respect to its own governing purposes. Thus when Mill tells
us that "human nature is so constituted as to desire nothing
which is not either a part of happiness or a means of hap-
piness," 9 it would seem that he did not mean to assert that
we desire what we desire. One is forced to conclude that in
such a passage, "happiness" or "pleasure" ( for Mill, of course,
they are equivalent) are not to be understood simply as
synonyms for "object of desire" or "realization of desire."
Surely, one supposes, Mill, with his awareness of human
ignorance and his fear of the uncorrected impulses of the
mob, did not wish to assert that in our dealings with others
what we should regard as intrinsically desirable is the
realization of their desires as they stand. Surely, as a con-
sequentialist who put his emphasis not on motives but on
their fruits,he did not mean to enjoin us to regard objects
of desire alone as intrinsically worthy of our respect and
solicitude. And yet, to our confusion and despair, we find
him saying almost in the same breath that "desiring a thing
and finding it pleasant" are not merely "two parts of the
same phenomenon," but "in strictness of language, two modes
of naming the same psychological fact." x So construed, it
would seem there is no difference whatever between ethical
hedonism and ethical voluntarism, and that psychological
hedonism is a vacuous tautology, not a theory at all. The
truth of the matter is that, by an inattention to a funda-
9 Utilitarianism,
p. 36.
ilbid.
mental ambiguity in the term "pleasure" and its derivatives,

Mill allowed himself to fall into a confusion which is fatal
not for his meta-ethics but for his normative moral philos-
ophy.
Now one sense, of course, Mill was quite within his
in
right in equating "desiring" and "finding pleasant"; both ex-
pressions, as Professor Ryle tells us, 2 do refer to inclination,
which is a sort of proneness or readiness to do certain sorts
of things on purpose. And in another sense, he is also justi-
fied in talking, as he so often does, about a "desire for pleas-
ure." But this latter expression requires that "pleasure" and
"desire" be distinguished and, indeed, contrasted. So under-
stood, "pleasure," as opposed to "desire," does not refer to
inclinations, but rather to certain distinctive feelings or
moods which we designate by such expressions as delight,
joy, contentment, the sense of well-being, and (in one
sense) satisfaction. But it obviously makes a tremendous
difference, for moral philosophy and for those who are en-
joined to base their deliberations on the assumption that
pleasure alone is desirable, which of these senses of "pleas-
ure" is intended. On one interpretation we are, in effect,
enjoined to contribute as much as we can to the delight,
joy, contentment, well-being, and satisfaction of ourselves
and others. On the other interpretation we are enjoined, I
suppose, to respect (or coddle) other people's inclinations
and wishes, to help them to get what they happen to be
aiming at. No greater ethical difference, I submit, can be
imagined. For my part, it is plain that after all it is the
former sort of thing toward which Mill's hedonism was funda-
mentally oriented. 3
2 Gilbert Ryle, The Concept Mind (London: Hutchinson,
of 1955),
pp.8 3 ff.
3 Much the same sort of difficulty appears when we turn to Mill's
discussion of "qualities of pleasure." The phrase means one thing if

"pleasure" is construed as a feeling word, but means quite another thing
if "pleasure" is taken as equivalent to "desire." And if "quality of
pleasure" is construed as "kind of feeling," a different thing is meant

by regarding some kinds of pleasure as "higher" than others than is
meant if, as Mill suggests, "quality of pleasure" means nothing more
than "decided preference." Again Mill is simply confused. And in any
I have dwelt again on Mill merely to illustrate once

again my contention that philosophical analysis is an indis-
pensable tool for normative ethics and that the clari-
itself,
fication and subsequent definition

is an indis-
of aims
pensable part of successful practical deliberation. Other il-
lustrious examples of moral confusion and of the use of
analysis toward its removal come readily to mind. The his-
tory of egoism, as everyone knows, is one long history of
equivocation, ambiguity, and vagueness. Its plausibility or
attractiveness for many thoughtless people has depended
largely, I think, upon these factors. When they are elim-
inated by such superb essays in clarification as the second
appendix to Hume's second Enquiry, the business of
normative moral philosophy is thereby immeasurably ad-
vanced.
ON INFERRING ETHICAL CONCLUSIONS

FROM FACTUAL STATEMENTS
We have now to consider the second part of Mr. Prior's
statement, namely, the wider thesis that a fallacy is in-
volved whenever we
attempt to deduce or claim to deduce
moral conclusions from statements of fact. I have already in-
dicated in what sense this seems to me to be plausible. I
shall now try to show in what sense it is mistaken and mis-
leading. Let me again emphasize that my reason for be-
lieving that moral conclusions may be inferred from state-
ments of fact does not in the least depend upon the tacit as-
sumption that moral conclusions are, after all, a species of
factual statement, or that the ethical terms occurring in the
conclusions are analytically synonymous with descriptive
case, to promote or support other people's preferences, when well-con-
sidered, is one thing; to promote certain "higher" qualities of feeling or
certain experiences of pleasure such as those involving the exercise of

our so-called higher faculties, is something else. All through Mill's
moral and social philosophy one can find a tension which arises pri-
marily on the fact that he had not sufficiently clarified what he re-
garded as intrinsically valuable. I do not say that the tension would
automatically disappear once the ambiguities in the term "pleasure"
were brought into the open. But I do say that only then could one
intelligently decide, as a putative hedonist, what one was really for.
)
or empirical predicates occurring in the premises. This is as

it may have already indicated elsewhere that I do
be. I
think that in certain contexts "good" and "right" tend to

acquire fairly well-defined descriptive meanings. But my
present argument does not depend upon this fact. Nor do I
wish, at this point, to cavil at the difficulties involved in
such ill-defined expressions as "ethical terms," "ethical con-
clusion," "nonethical term," "empirical or natural predicate,"
"nonethical premise," and the like. The difficulties implicit
in the ordinary use of these expressions in contemporary
ethical theory I have also pointed out elsewhere. 4 In his
essay, "A Finitistic Approach to Philosophical Theses,"
Morton White has shown how tenuous is the thesis of an-
tinaturalism in its ordinary "infinitistic" formulations. 5 There
will be some naturalists who may choose to avail themselves
of Frankena's and White's criticisms of the question-beg-
ging assumptions and imponderable infinitistic theses in-
volved in the usual arguments provided by antinatural-
6
ists in their attack upon the naturalistic fallacy. I do not wish
to avail myself of these criticisms, just though they are, for I

do believe that, when all the question-begging assumptions
and the infinitism have been removed, there remains a sense
in which the naturalistic fallacy really is a fallacy.
In spite of this, I believe that under certain conditions it is
entirely proper to infer (not deduce), ethical conclusions
from factual premises. (The quotation, on page 53, from
Professor Robinson provides an interesting example of this.
And the reasons why some philosophers have supposed that
a fallacy is involved in such inferences are (a) the tenacious
hold which descriptive models have upon all of us in all
our interpretation of language, and, more important, ( b ) the
ingrained tendency to regard the processes of validation ob-
taining the formal logic and in inductive reasoning as the
4 See my "Evaluation and Obligation: Two Functions of Judgments

in the Language of Conduct," Journal of Philosophy, XLVII, No. 1
(Jan. 1950), pp. 5-7.
5 See M. G. White, "A Finitistic Approach to Philosophical Theses,"
Philosophical Review, July 1951, pp. 307-11.
6 See ibid.; also Frankena, op. cit.,
pp. 472 ff.
only processes in which questions of validity can ever arise.

So tenacious are these assumptions that, when certain philos-
ophers get even the vaguest inkling that there may be in-
formal contextual standards of relevance in art and morals,
they leap immediately to the absurd conclusion that a work
of art is a deductive system or that a moral code is one of
the social sciences. Perhaps the most fantastic example of
this sort of thing in recent literature may be found in Pro-
fessor James Feibleman's Aesthetics. According to him, the
is simply a deduction of the logical
act of artistic creation
consequences implicit in the artist's initial idea. 7 He says, in
so many words, that the relations holding between a theme
and its variations are not merely analogous (in some sense)
to, but identical with those which hold between the axioms
and theorems of a deductive system. Such a view, you may
say, is absurd and I agree. It is logicism gone mad. But I
am convinced that nestled at the heart of the absurdity is a
genuine insight, if he only knew how to make use of it.
This insight, however, is concealed from its author by virtue
of his unconscious addiction to the idea that validity and
relevance are notions which are exemplified only in logical
systems. The insight is this: there are relations of relevance
—some people them relations of "propriety" or
call "fitting-
ness"—which hold between the parts of a successful work of
art.
8
And in the same way, although it is absurd to say
that ethical conclusions can be deduced from factual prem-
ises (since here we are dealing with propositions, the ab-
surdity is less palpable), there is still an insight of sorts
concealed within the absurdity. For there are conditions of
relevance which permit, in certain circumstances, the infer-
ence of significant ethical judgments from statements of fact.
Some of these I will mention shortly.
Meanwhile it must be remarked that in recent years
moral philosophers have been more prone to see the ab-
7 See James K. Feibleman, Aesthetics ( New York Duell, Sloan & :
Pearce, 1949), pp. 12 ff.

8 See my "The Concept of Relevance in Aesthetics," Journal
of
Aesthetics and Art Criticism, Dec. 1949, pp. 152-61.
surdity than the insight within it. And this is due, again, to a
subtle addiction to descriptivistic attitudes implicit in the
above-mentioned notion concerning validity. Thus, for ex-
ample, although Professor Charles Stevenson is perhaps more
sensitive than any other philosopher to the reality of per-
suasive definition, he himself falls into it when he refuses,
for no reason sanctioned by ordinary language, to accept the
possibility that there are any "rational methods" other than
those of formal logic and science. Although he freely allows
the right of moralists to use, inter alia, what he regards as
rational methods when they happen to be appropriate for
the purpose of "irrigating" ethical judgments, he nevertheless
insists there are no criteria of validity with respect to ethical
disputation as such. But why should he fear lest the no-
tion of "validity" be extended so as to include forms of in-
ference which are neither demonstrative nor inductive? As
he himself wisely says, "When an inference does not purport
to comply with the usual rules, any insistence on its fail-
ure to do so is gratuitous." 9 And yet he maintains, to my
mind quite unconvincingly, that it is "wholly impracticable
and injudicious" (sic) to sanction a definition of validity
which extends its usage beyond its applications to logic and
to science. 1 Apart from a tenacious desire to reserve the
emotive meaning of such expressions as "rational" and "valid"
for processes of reasoning involved in formal logic and in-
ductive science, what is there to commend Stevenson's posi-
tion? Again, why must it be misleading to say that "validity,"
not in special "philosophical" senses but in normal common-
sensical applications, is a normative as well as perhaps (in
some uses) a descriptive term? Surely it is so. To say that
"x is invalid" is, conform to certain
in effect, to say "x fails to
accepted conditions; do not accept x." In the realm of logic,
its function is to control and direct belief. In the realm of
ethics, it is not only to control belief, but also to control and

direct practical second-level attitudes.
I submit that Professor Stevenson is still, in an extremely
9 Stevenson, op. cit., p. 153.

1 Ibid.,
p. 154.
subtle and involved way, under the spell of descriptive

and models in his approach to nonscientific forms of
logical
reasoning, even though he is quite free from such models
in his analysis of nonscientific terms. To be sure, he does not,
like Perry, make the mistake of interpreting moral disputa-
tion simply as a variety of scientific and logical inference,
albeit with a subject matter that cuts across the special
sciences. On the contrary. Yet he does appear to regard
science and logic as the proper models in regard to all ques-
tions of relevance and validity. And it is because of this, I
believe, that he is forced to conclude that there is no such

thing as validity in morals except by accident, when purely
factual or logical issues are momentarily in view.
In principle, of course, he on his view, to dis-
is free,
approve those who employ arguments in the proc-
"invalid"
ess of supporting ethical judgments. That is his privilege as
a moral being. But as an analyst he is bound to regard such
arguments, from the standpoint of moral discourse itself,
as entirely within the proprieties. Thus although he may, as
an individual, be against certain forms of "irrationality" in
ethics, he is obliged, as an ethical theorist, to accept them as
integral parts of moral persuasion and argument.
The issue here, I realize, is exceedingly delicate —so much
so, indeed, that I am not altogether sure that I too have not
been unintentionally misleading. Perhaps, in the end, all one
can do is to go on indefinitely correcting the misleading im-
pressions of one's own preceding remarks. This may, in fact,
be all that writing a book in philosophy amounts to. I must
state, in any case, that I get the strongest sort of impression
from Professor Stevenson's writings that he is almost as
much concerned to protect science and logic as the only
valid methods of argumentation as he is to free ethics itself
from bondage to a false god of rationality.
Now I agree once and for all that there are no formal
logical rules by means of which one can deduce the ethical
proposition "x ought to be done" from any combination of
purely factual statements. What I do maintain is that, ac-
cording to ordinary usage, it is entirely permissible to infer
ethical conclusions I should now like

from factual premises.
to support this contention with some examples. Suppose that
it could be shown that a certain act would cause another per-
son unnecessary hardship or suffering; I think that any nor-

mal person in our society would regard this as a good, if
not sufficient, reason for inferring that, other things remain-

ing equal, the act in question ought not to be performed.
Again, suppose itcould be shown that the fulfillment of a
certain promise would probably cause the person to whom
one made it to destroy himself; here also, I think that normal
persons would, perhaps reluctantly, conclude from this that
the promise ought to be broken. Other examples come to
mind.
1 conclude from this that, however difficult they may be
to specify, there are nevertheless broad principles of rele-
vance or valid inference in moral discourse which enable us,
in certain circumstances, to infer ethical conclusions from
nonethical premises. But I do not in the least wish to imply
by this that the ordinary laws of logic should be amended
or broadened. Such laws have no immediate application
to the kinds of inference in question. My contention is
merely that within the universe of discourse called "moral"

or "ethical," certain types of inference are viewed as reason-
able, others not. Nor do I wish to say that moral judgments
may be "logically derived" from nonethical statements of
fact. 2 I think, nothing is gained from such an unnecessary
and really misleading extension of the expression "logical
derivation." All that needs defending is the thesis that
moral reasoning has its own proprieties which, while cer-
tainly not written into the starry heavens above, are at least
constant and extensive enough to enable the members of a
given civilization to distinguish a good reason from a bad one.
2 Cf. Hampshire, op. cit., pp. 470 ff. Let me state here that, although
apparently different sorts of stimuli have given rise to our respective
reflections on moral philosophy, I find myself in agreement with much,
although not all, of Mr. Hampshire's admirable essay. In my judgment
the most stimulating thinking about ethics to be found in contemporary
philosophy is being done at Oxford by Mr. Hampshire and his col-
leagues.
[IV]
Levels of Moral Discourse
THE POINT OF VIEW OF THE PRESENT ESSAY

As a matter of principle most contemporary moral philoso-
phers pay lip service to the diversity and complexity of
human problems. One might reasonably assume, therefore,
that they should regard it as axiomatic that the forms of
discourse most intimately related to these problems would re-
flect their diversity and complexity. Yet when we survey the
prevailing contemporary moral philosophies, we find that all
or nearly all of them are monistic in their approach to the
meanings of so-called "ethical" terms and reductivistic in
their treatment of the roles of so-called "ethical" judgments.
If they succeed momentarily in avoiding the "one and only
one meaning fallacy," they do so only to fall directly
into the "one and only one function fallacy." Even when
there is a passing acknowledgment of the "ambiguities" of
such terms as "good" and "ought," there is usually an im-
mediate narrowing of the subject for analysis to some "essen-
tially" or "characteristically" ethical sense of these words.
Essences, however, always follow our interests, and where
interests differ, what seems essential to one will appear
merely accidental to another. Because of this it is not sur-
prising that there should be disagreement among moral
philosophers as to what is characteristically, or essentially,
ethical in themeanings of ethical terms and in the roles of
ethical judgments. The unfortunate consequence of this, as
one might anticipate, is an interminable debate among the
various ethical theories which is profitless because it is in-

soluble.
The only way to resolve such controversies, which merely
block the progress of moral philosophy toward a more com-
prehensive grasp of its problems, is simply and resolutely to
ignore all the essentialistic and reductivistic questions upon
which they depend and to proceed at once to a detailed

and unpolemical examination of the several levels upon
which practical discourse proceeds. The resulting analysis will
not be classifiable under any of the main "types" of ethical
theory that now prevail. It will be neither a cognitive nor
an emotive theory, though it will be prepared to acknowl-
edge both the many cognitive and the many emotive aspects
of the language of conduct. It will be neither subjectivistic
nor objectivistic, though it will attempt to show that moral
discourse has, and must have, both subjective and objective
phases. It will acknowledge the importance of reasoning in
conduct, but it will also recognize its limits. It will see that,
while there such a thing as "justification" in ethics, there
is
is no one form to which all the justifications occurring within
moral discourse can be reduced. In short, such a theory will

be pluralistic in orientation, and hence fundamentally op-
posed to all of the "type" theories whose exposition and
defense occupy so large a place in the current literature. It
will not, of course, oppose reduction indiscriminately; but it
will be prepared from the outset to acknowledge irreducible
differences in meaning and irreducible differentiations of
function. Nor will it be misled by the countercharges of
"eclecticism" or "lack of system." "Logical" or "analytic" con-
nections of meaning are not the only relations among con-
cepts or judgments that require explanation, nor is there
only one way in which "system" can be achieved.
Such, at any rate, is the point of view of the present
essay. Its purposeis to show that there are at least four
upon which such terms as "good," "right,"

distinctive levels
and "ought" are employed, and to explain the roles of judg-
ment on each of these levels.
For purposes of identification I shall speak of these levels
—
Levels of Moral Discourse 67
respectively as (1) the "expressive-evocative" level, (2) the

"moral" level, (3) the "ethical" level, and (4) the "post-
ethical" or "human" level. Such labels are perhaps arbitrary.
They are here defined exclusively with reference to specific
contexts of moral discourse. To dispute about them in any
intelligible way is to argue the adequacy of the analysis of
a particular context or level. By which name we call the level
is a question of secondary importance.
Before proceeding to a detailed examination of these
levels, however, several preliminary remarks need to be
made. Without them any schematic survey of the sort pro-
vided here must inevitably prove misleading. In the first
place, then, any moral argument which goes on at any length
is likely to proceed on more than one level. As questions
become more probing or more urgent, more and profounder

resources of the language of conduct will be brought into
play. This means that the context of moral discussion is, or
tends to be, a shifting context. Second, the nature of moral
discussion being, as it is, practical both in intention and in
effect, the connections between the successive levels must
be understood primarily in pragmatic rather than in "logical"
terms. Such systematic relations as may be discerned in the
shifts from level to level will be functional rather than
deductive or evidential in character. But, third, this should
not be taken to imply that on a given level there are no criteria
of relevance or validity. On at least two levels 2 and 3 —
such criteria do obtain, as I shall presently show. Only at the
extremes do we pass beyond the bounds of "propriety" or
open sea of individual feeling or human
"rationality" to the
aspiration. Here may be relevancies and irrele-
also there
vancies, but they are of a different sort altogether, and can
be determined, if at all, only by appealing directly to the
irremovable needs of the individual agent. This is why, as
we shall see in more detail later, there is in moral dis-
course a variety of modes of justification. Some are in an
intelligible sense "objective"; others, whether we like it or
not, are necessarily "subjective." The differences between
them can be explained only with reference to presence or
absence of intersubjective "rules" or "principles" to which one

may confidently make appeal in carrying out a justification.
An objective justification, since it involves an appeal to in-
tersubjective rules, is immediate claims of
indifferent to the
personal inclination or preference. Indeed, it is on this basis
alone that we can make out a clear distinction between the

"desirable" and the "desired," or between "apparent goods"
and "real goods." But in the end any justification which is
practical in intent must provide a justification to interest.
So far it must be subjective or it has no relevance to action.
Such a justification cannot in the end be indifferent to the
demands of inclination, since here there is nothing in the
last analysis to appeal to save human passion itself. Passion
is prior to rules, if, within limits, it is also governed by
rules. Both forms of justification have their place in ethics.
The danger lies in failing to distinguish them.
THE EXPRESSIVE LEVEL

Our initial responses to any situation are likely to be of the
unreflective, stock variety. We see something and like it; we
hear something and dislike it; we think of someone and
are at once attracted or repelled, we know not why. Unless
there is something unstable or ambiguous in the situation
which requires reflection or deliberation, we are usually
content, if we make any comment at all, simply to express
our passing feelings of favor or disfavor. "Bravo!" we cry as
the curtain goes down. day for a picnic!" we
"What a fine
exclaim as we step out on the porch on a sunny day in June;
"What rotten luck that the attic has to be cleaned out."
"Good play!" we shout as the first baseman on the home
team fields a line drive headed for the bleachers. Such ex-
pressions of spontaneous pleasure or displeasure serve merely
to vent our emotions; they pose no problems nor call for
any reply. They are immediately construed by others for
what in fact they are, namely, as conventional expressions
of personal feeling. To challenge them or to raise questions
of "truth" or "validity" with respect to them would be
pointless. They do not solicit agreement or invite a reply.

Such uses of "good," "bad," and their equivalents have
received much, perhaps too much, attention of late, due to
the advent of the emotive theory. It would be a mistake
to disregard them, however, or to fail to recognize that, even
at this most elementary of the levels upon which the language
of conduct is employed, there are more complexities than
are usually noticed. Failure to discriminate them can cause
trouble later on at the levels which perhaps interest us more.
A full account of the expressive level would have to dis-
tinguish at least the following: (a) the conventional use
of words as vehicles for the expression of emotion; (b) the
varying degrees of intensity and rhetorical force of such
expressions; (c) the expressive or "venting" relation holding
between the speaker and the expression itself; (d) the
symptomatic relation in virtue of which the expression func-
tions as a natural sign to an interpreter; (e) the incitive
or rhetorical effect of the expression upon an interpreter, in
virtue of which we speak of a relation of communication
between him and the speaker; and (/) the intentions of the
speaker in thus giving vent to his emotion. Here already it
is clear that several modes of "meaning" may be distinguished.
And in the case of d at least, it is plain that, as early as

the expressive level, "cognitive" meanings begin to emerge,
even though nothing is "designated" or "named" as such.
Even here, therefore, a pure emotive theory would not
account for all the facts.
THE LEVEL OF MORAL RULES

At the purely expressive level, as such, no question of justi-
fication can possibly arise. Nor can there be any problem

whether what is directly responded to or reported as "good"
or "bad" really is so. Here everything is as it appears to be.
Such diversity of expression as may arise signifies nothing
—
more than the venting of contrary or perhaps merely differ-
—
ent emotions. To ask who is right in such a context or
whether the expressions "good" or "bad" are really appro-
priate to the objectstoward which they are directed would

—
be senseless or rather, which is closer to the facts of the
matter, it would be to shift the discourse at once to another
level.
The first level upon which serious questions are asked and
serious answers given in ethical terms I have called the "moral"
level.Here questions for the first time begin to emerge: "What
ought I to do in this situation?" "Is this object that I admire
so much, really good?" "Is it really worth having?" Here, in
short, there now appears a problem of conduct and a
problem for appraisal and ultimate decision. On this level
two sorts of utterance are involved in attempting to justify
one's answer to such a problem: (a) factual appraisals of
relevant means and consequences and (b) rules or proce-
dures in relation to which alone the moral relevance of such
appraisals can be established. Frequently one of these goes
unmentioned. When the facts are sufficiently plain, atten-
tion may be directed exclusively to the interpretation and
ordering of the relevant rules. At other times there may be
no question concerning the meanings, order, or application
of the rules, but only a problem as to the facts themselves.
Moral discourse, being practical and frequently urgent in in-
tent, is full of such ellipses and elisions. The justifying reasons
for many of the things that we ought to do go without saying.
Because of this, so discerning an analyst as Prichard could
argue very plausibly that our particular obligations are
really self-evident. 1 That they are really not so, however, is
sufficiently indicated by the facts of moral perplexity and
disagreement. It is always legitimate, if often fatiguing, to
ask "Why?" when a particular moral judgment is proposed.
And in such a case, if we are reasonable, there is no escape
from an appraisal of the relevant facts and, if necessary, an
explicit appeal to the appropriate rules.
The tacitness of so many of the understandings governing
our moral deliberations has caused many philosophers to ig-
1 Cf H. A. Prichard, Moral Obligation (London: Oxford University

.
Press, 1950), pp. 1-18.

nore one or the other of the roles of factual reasoning and

rules inmoral discourse. The neglect of either, however, in-
variably results in a form of irrationalism. When, on the one
hand, the role and relevance of factual premises in our
inferences to moral conclusions is overlooked, the distinction
between what is and what ought to be becomes a total
diremption, with the consequence that "insight," as in the
case of Prichard, or "sentiment" as in the case of Hume (in
some passages), is substituted, at the wrong place, for ra-
tional reflection. When, on the other hand, the role of moral
rules is ignored, as in the case of Stevenson, the whole con-
ception of validity or relevance in ethics tends to lose its mean-
ing altogether, with the result that "justifying" reasons of all
sorts collapse at once into "exciting reasons."

Moral rules still govern the course of our factual reasonings
in ethics, for all the fact that, unless an issue of interpreta-
tion or precedence arises, there is usually no need to refer
explicitly to them. But in the last analysis, they alone deter-
mine what factual reasons are to be accepted as relevant.
Not just any facts or consequences have bearing upon a
moral problem. The besetting weakness of the instrumental-
ist account of moral deliberation is that it provides no
criteria as to the relevance of the consequences upon which

it enjoins us to reflect. Hence, in spite of their desire to
render moral deliberation "scientific," the instrumentalists

neglect the very considerations which might conceivably turn
science to a moral use. The emotivists are more candid. They
hold that there are no criteria of relevance or validity in
moral deliberation at all. Any statement of fact, in their
view, is ipso facto appropriate to the extent that it proves
modifying the attitudes which we
efficacious in reinforcing or
wish to But in practice, moral reasoning is
fortify or change.
not so deucedly liberal. For better or worse, it presents to
us certain standards of propriety to which we regard our-
selves as bound. Some reasons it acknowledges to be "good,"
if perhaps, in some cases, insufficient; others it holds to be
"bad," despite their rhetorical effectiveness. On this level, a

good reason is one whose relevance to the problem at issue
isdetermined by the rules of the communal code itself. A

bad reason on this level is one which is merely "exciting."
Now the rules of moral codes vary considerably in num-
ber, clarity, flexibility, and well-orderedness. In "closed" soci-
eties they are likely to be many, specific, rigid, and well
ordered. The studies of such anthropologists as Malinowski or
Ruth Benedict amply confirm this fact. In more open societies
rules are usually fewer in number, highly general, capable
of indefinite reinterpretation, subject to exceptions, and
rather vague in order of precedence. When a problem of
interpretation arises, as it frequently does in such a "system,"
the individual is back upon another
frequently forced to fall
concept, which is nonetheless important because it is even
vaguer than the procedures between which it is required to

adjudicate. This is the notion of "good sense," "normality,"
"reasonableness," or "competence." To suppose, however, as
some writers have done, that this principle does all the pro-
cedural work in moral deliberation is as serious as the anal-
ogous supposition would be in regard to the law. We must, to
be sure, be "sensible" if we are properly to resolve our moral
perplexities. But this is not to say that good sense works in
a vacuum without the guidance of more explicit rules of
conduct.
In this connection it may be useful to compare my views
with those of Stephen Toulmin and John Rawls. In his dis-
cerning review of Toulmin's An Examination of the Place of
Reason John Rawls makes the following remark in
in Ethics,
"The point is that a reason is any
criticism of his author:
consideration which competent persons in their reflective
moments feel bound to give some weight to whether or not
they think the consideration sufficient in itself to settle the
case." 2
I should myself amend this to read: A reason (in
ethics) is any consideration, including those covered by
acknowledged rules of the moral code, which "competent"

persons in their reflective moments feel bound to give some
weight to whether or not they think the consideration suffi-
2 Philosophical Review, LX, No. 572-80.
4,
cient in itself to settle the case. I agree with Rawls that

Toulmin's account of moral reasoning is But
far too rigid.
I agree with Toulmin that there are moral rules, although it
seems to me that he supposes them to be more objective,

definite, and inflexible than they are. My own view, in
short, is somewhere midway between that of Toulmin and
that of Rawls.
In the spheres of logic and factual explanation there is, or
so we like to suppose, only one set of correct or valid
principles of reasoning. 3 But in morals, unfortunately, there
just isno one set of universally valid principles of delibera-
tion to which all peoples, regardless of cultural heritage, are
in conscience bound. In saying this, I am not preaching moral
relativism; I am simply stating a fact. It is a fundamental
theoretical blunder to treat the particular procedures of moral
deliberation current in our own culture as paradigmatic for
morality in general.
Here isanother point on which Stephen Toulmin seems
to me to be somewhat unguarded. His emphasis upon utili-
s I take no great pleasure in the thought, but this statement may be

more debatable than some of my more sanguine empiricist allies would
admit. The issues which it raises, however, cannot be argued here.
Fortunately, they do not immediately affect the problems with which
we are here concerned, although I must warn my colleagues against the
tendency to identify "good sense" in the sphere of factual assertion with
the criteria which they themselves accept as correct or proper. Cf. Pro-
fessor Herbert Feigl's essay, "De Principiis non disputandum ?" in. . .
the volume Philosophical Analysis, edited by Max Black (Ithaca, N.Y.:

Cornell University Press, 1950). Says Feigl: "The vindication of the
principles of meaning and knowledge is so trivial precisely because,
given the purposes of language and knowledge, there are no genuine
alternatives for fulfilling them. But we do know of alternative systems
of moral norms." (p. 143). I wish Feigl were right. He certainly con-
vinces me that given the uses I have, if not for language, then for
"knowledge," there are no alternatives for fulfilling them but the princi-
ples he has in mind. Unfortunately, mine are not the only uses that have
prevailed in the history of thought. Nor are his criteria, which are
— —
broadly speaking my own, the only ones to which reflective persons
have appealed, even in our own tradition. I am curious to know how
Feigl deals with the "thoughtful" persons for whom Thomas Aquinas
is the spokesman. I condone their standards no more than Feigl does,
but I think that they have different purposes for and different principles
of "knowledge" than he and I do.
tarian arguments which show that actions not otherwise

prescribed as duties will cause other members of the com-
munity some inconvenience, annoyance, or suffering is, I
think, largely correct for "our" system. That it is a correct
characterization of other codes seems to me very doubtful.
This mistake, if such it be, is, I think, a consequence of the
tendency to argue from the general social effect of morality,
which is usually to prevent or reduce conflict, to the nature
of the types of reasoning that govern it. Even the edicts of a
totalitarian system prevent conflict in many cases, but I
doubt whether anyone would care to argue that this is more
than a by-product.
We must not assume, therefore, simply on the basis of our
own conviction, that the proper end of moral reasoning
should be the harmonization of actions, that the principle of
harmony is everywhere the standard of validity in moral
reasoning. The initial plausibility of the assumption is, I
admit, somewhat increased by the fact that a common social
effect of morality is, as a rule, to prevent conflict and insure
order. But this is, within limits, the usual effect of any rule-
governed activity or institution. It does not follow that it is
to be acknowledged as a principle of "right reason" in moral
deliberation. Again, such humanistic principles as "harmony,"
"adjustment," and "least suffering" are implicit to a greater
or less degree in the codes of many "civilized" peoples.
But however desirable they may be from our point of view,
they are still not universally definitive of what it is to be a
moral rule or principle. The conclusion cannot be avoided,
therefore, that at present there just is not one set of proce-
dures or rules for the validation of moral reasoning to which,
regardless of cultural differences, every person is "objectively"
bound. On the other hand, it is not necessary to infer from
this, as some writers have done, that the very concept of
validity has no meaning when applied to moral arguments
and deliberations.
What I here call "moral rules" at once specify certain types
of behaviorwhich ordinary non-deviant persons within a
given community would approve and which demand that
the addressee, in so far as he is "normal," likewise approve

and, if appropriate, act accordingly. 4 The moral judge, so
understood, functions primarily as a middleman or agent.
He voices the claims of society but is not the primary source
of their moral authority. This does not imply, of course, that
the moral judgment does not express the sentiments of the
judges. It may; but though, as normal persons, we tend to
adopt the attitudes which we express in our moral judgments,
this fact is not essential to their characteristic meaning or role.
It is the social press of "morality" itself which disposes us to
present its claims, not the personal interests we may muster
in its defense. This is reflected in the fact that at the
phenomenological level morality normally appears to us as a
constraint, not as an expression of our inclinations.
THE LEVEL OF ETHICAL PRINCIPLES

Ordinarily, we do not need to proceed beyond the "moral"
level in our deliberations. But answers of the sort hitherto
considered do not necessarily put an end to the ethical ques-
tions that may relevantly be raised. Occasionally one is
obliged to ask whether an action which is prescribed by ex-

istingmoral rules really is right and whether, therefore, one
ought to continue to obey them. When pressed in a certain
way, the effect of such a question is to throw doubt upon the
validity of the rules themselves. And in that case, there is
usually no alternative to a fundamental reconsideration of
the whole moral code.
Such questions have many causes. It may be that the moral
rules conflict, or that a consistent adherence to them would
result in general inconvenience or suffering. It may be that
they run too persistently against the grain of human need
or inclination. It may be that changing social conditions
render them inapplicable or inadequate for the adjudication of
communal disagreements. We must distinguish, however, be-
tween the various causes which may animate what I here
4 For a more detailed defense of this characterization see "A Plural-
"
istic Analysis of the Ethical 'Ought.'
distinguish as "ethical" criticism and the procedures which

give it its status as a valid mode of practical reasoning.
Several aspects of ethical criticism require comment. In
the first place, it must be constantly borne in mind that
when we question the propriety or fitness of a moral code
our question remains practical, both in intention and in
effect. Ethical questions are never purely theoretical or
speculative. But there are many ways of raising practical
questions. When they are raised in ethical terms, the effect,
first of all, is to place the questions on a level of imperson-

ality which requires the subordination of personal bias or
preference. It is their function to establish a mood in which
the particular moral rule or the moral code as a whole is
considered impartially or, as we say, "objectively," without

regard to our own inclinations or benefits. In this guise, the
incitive or normative effect of ethical terms appears to
consciousness as something which is independent of and even
in opposition to "interest." A
second characteristic effect of
this use of ethical terms tendency to "frame" or set
is their
apart the questions and answers in which they occur from
ordinary practical deliberations. The language of ethics is,
among other things, a ceremonial or ritual language. Our
"lay" deliberations and appraisals may be fully as important
for our ultimate well-being as those couched in ethical terms.
In certain situations they may be even more intensely in-
citive. What they lack is the distinctive "aura" of impersonal
authority which sets ethical judgments apart from the ruck
of practical demands. "Would it be beneficial to everyone
considered if the moral code were revised?" has nothing
like the distinctive authority of "Would it be right to revise
it?" or "Ought we to alter the moral code?" Here the only
illuminating comparison is with other ceremonial languages

such as those of religion or the law. Each of these institutions
has its own characteristic rituals, each of which is incitive in
a somewhat different way. Each has its own characteristic
"style" which signifies a different sort of press and different
trains of reflection. The ceremonies of the law are impersonal
and authoritative, but they carry with them the suggestion
of ulterior punishment or confinement. They are inevitably

threatening in their effect in a way that the language of
ethics is not. On the other hand, the distinctive ceremonies of
religion, although impersonal, are primarily devotional in
character; the press which they evoke is one of worship,
prayer, and meditation. Their function is discharged when
they have evoked the proper moods and feelings. The atti-
tudes aroused by ethical terms on the level of discourse
which we are considering, however, are not worshipful but
critical, not meditative but, by intention at least, practical.
Perhaps the most distinctive feature of ethical discourse is

its so-called "autonomy." It makes no promises of future
benefit to the individual; nor are its principles justified by

an ulterior consideration of expediency. If such a justifica-
tion is made, it has nothing directly to do with ethical criti-
cism as such. Indeed, it is precisely the effect of ethical
terms to short-circuit questions of this sort, to shame them,
in effect, out of countenance. To ask whether a proposed
reform of the moral code will make one happier or more
prosperous is to declare one's self so far immoral, and
hence subject to moral censure. But the only "threat" im-

plicit in moral judgment is the impersonal disapproval which
is embodied and projected in the judgment itself. Psycho-
genetically, the "source" of the authority of the language of

morals is closely connected with the sanctions of parental
approval and disapproval. Gradually, however, "ought"
and "right" become incitive substitutes, not merely for the
parent, but for the social group. And at the level of ethical
criticism they appear to voice a claim which is wholly ideal
and universal. The point is, however, that in fact there
is nothing apart from the attitudes which these terms evoke
in one's self to substantiate or enforce the claim. One may give

"reasons" in support of this or that demand for a change in
the moral code. But at last one can only justify such reasons,
from an ethical point of view, by appealing to ideals or
standards which themselves establish what we mean by an
ethical reason. To require their justification is simply to go
beyond "ethics" altogether.
Now, as Hume understood, more clearly, perhaps, than any-

one else, the interests which support "morality" are many.
Custom, habit, sympathy, fear, the love of reputation, and a
thousand other motives, together with the myriadic nonmoral
"reasons" that reinforce them, combine to secure moral action
upon a far more solid base than would be possible were
we to rely exclusively on moral justifications. In the same
way, a thousand and one nonethical motives and reasons
may support the ethical reasonings in the light of which a
reform of the moral code may be demanded. There is no
need to impugn such nonmoral considerations or to deny
them their place in the sphere of practical reasoning and
deliberation in general. Indeed, when such ulterior supporting
considerations are long absent, morality tends to die of sheer
attrition. To acknowledge this, however, is not even by im-
plication to blur the distinction between ethical and noneth-
ical reasons or between ethical principles and individual
preferences.
The same sort of point is frequently made in regard to
questions of logic or questions of fact. In trying to convince
someone of the truth of a certain statement, we often find it
necessary to introduce supporting reasons that are not strictly

relevant from a logical point of view. It is sometimes actually,
if unfortunately, the case that people will not listen to
"reason." When this happens, we have to employ other
methods of persuasion. But this does not in the least mean
that we confuse a "good" reason in logic or in factual reason-
ing with one which simply happens to work. Nor does it mean
that we are impelled to question, in their own spheres, the
autonomy methods of logic and inductive
of the validating
science. If someone challenges these methods or asks us to
supply a better reason than they afford in answer to the
questions they enable us to answer, we would simply not
know how, within logic or science, to make a reply. Prag-
matic "justifications" of the rules of inquiry doubtless have
their function, but they do not determine in general what a
"valid" inquiry is.
The same thing is true in the case of ethical appeals. One

may question, if one must, the "justice" of the rules governing

the duties of wives or mothers in our society. But if one
should question the legitimacy of the standard of justice itself,
in virtue of which alone a question becomes ethical, there

is simply no way of answering without moving outside of
ethics altogether. Someone may ask, "Shall I accept ethical
standards of criticism?" if by this he wishes to challenge the
procedures of ethics as a whole. But he cannot expect an
answer to which it would be possible to ascribe the character-
istic of "validity." At the ethical level, "ought" and "good"
are simply not appropriately used save in connection with

such principles as "justice" or "least suffering." Their whole
function here is to "put us in mind" of such principles. Both
are parts of an indivisible level of discourse, whose bound-
aries are set by the interlocking relations of such terms
themselves.
How many justifying principles or rules of relevance there
are in our system is very hard to say. One such principle is
that of "benevolence" or "least suffering." It is always legiti-
mate to question the legitimacy of a moral rule when it can

be shown that continued adherence to it would cause un-
necessary or undue hardship or misery. It is likewise proper
to ask for a reform of the moral code when it can be shown
that continued adherence to it would result in some funda-
mental indignity to the "persons" of certain individuals. On
the other hand, it is not legitimate to demand a modification
of the code on the ground that adherence to it would cause
personal inconvenience to ourselves. Nor is it valid to demand
a change on the ground that the drift of history is against it.
In my judgment, there is, in our system, no one principle of
ethical criticism to which alone we may properly make
appeal in justifying a demand for moral reform. Various at-
tempts have been made to reduce the principles of welfare
or least suffering to those of justice, equality, and respect for
persons, or vice versa —always, I believe, without success.
Other have been made to justify the one group in
efforts
terms of the other. Hume and Mill, for example, were able
to show, within limits, the benefits which accrue to society
as a whole from a strict adherence to rules of justice. But

they usually avoided "hard cases" when so doing. The benefit
to society as a whole never provides a conclusive ethical
reason for acting unjustly. Nor does the justice of an act which
would cause irreparable harm to society as a whole certainly
might be supposed that here we
establish the act's validity. It
must appeal to some vaguer but more general standard of
"reasonableness" or "competence." But in this case such an
appeal would be of no avail if, as I suspect, most reasonable
men, faced with the necessity of choosing between the two
sorts of principles, would find no final or certain answer as to
which is preferable.
It is precisely at such a point as this that ethical tragedy
occurs. When we reflect upon the essential tragedies of
Antigone or Job, it is no "fatal flaw" in the individuals in-
volved that lies at the heart of the tragedy but, as Hegel
saw, a fundamental and inescapable conflict of ethical
principles, both of which are "right." In both cases adherence
to one principle of right action involves the protagonist in
head-on collision with another. The conflict cannot be avoided,
but it is ethically insoluble. Hence the tragedy. Wherever such
tragedies as Antigone may occur, there is a plurality of basic
ethical principles. Indeed, the very possibility of tragedies of
this sort is itself an index of such a plurality.
In a strictly ethical appeal, then, such principles as those
of justice and least suffering define the relevant applications
of the ethical "ought," and it in turn sets them apart as the
distinctive rules of the ethical "game." Together they consti-
tute a universe of discourse which is so far autonomous and
self -justifying. In this sense, even in "open" societies, morality
is a closed system in a way that the laws of politically
organized society are not. Laws are not, at least in open
societies, self-justifying. The authority of any moral system,
however, is grounded in the very terms of morality itself.
When we "criticize" a particular moral code it is usually on
grounds that are themselves ethical.
One can, of course, go "outside" morality altogether and
question its "value" from some ulterior point of view. Or one
may propose, in the manner of Bentham, to dispense with

"morals" altogether and deliberate exclusively in terms of
consequences without regard to their moral worth.
felicific
But such a question or such a proposal has no more place

within ethical criticism as here conceived than analogous
questions or proposals would have within the "games" of
logic and science. To propose that science adopt "agreeable
consequences" as a criterion of truth would, in effect, be
a proposal to abandon science altogether for something else.
And in the same way the proposal that morality should sub-
mit to some "higher" or "more fundamental" authority in
justification of its rules is tantamount to the suggestion that
ethical discourse be replaced by some other form of delibera-
tion.
One criticism that has frequently been leveled against
ethical principles of the sort in question is that they are
"empty." To a person faced with the problem of making
up his mind what he ought to do in a particular situation,
such principles will inevitably appear trite, empty, and
vague. This, as Mill recognized, is just as true of the prin-
ciple of utility as it is of Kant's categorical imperative. Such
principles are principles of criticism by means of which
lower-order moral rules are "Secondary" rules, as
justified.
Mill called them, are not to be replaced by the greatest
happiness principle. In attempting to justify particular actions
it is to "secondary" rules rather than to the latter principle
that we must appeal. What must be recognized here is

simply the difference of level to which moral rules and
ethical principles apply.
To regarding respect for persons on
criticize the principle
the ground that "empty" would be analogous to a
it is
criticism of the rules of inductive inference which argued

against them on the ground that no substantive law of
nature could be deduced from them. Such criticisms are
based upon a complete misunderstanding of what such
principles are designed to accomplish. As we move up the
ladder of moral criticism from justification to justification
and from subordinate rule to superordinate "procedure," we
also move from attitudes that are relatively determinate

to others which are relatively indeterminate. The former
are directed to the solution of particular problems of conduct
or concerned with the realization of particular goals. The
latter,on the contrary, are directed rather to the organization,
regulation, and correction of lower-order attitudes. Second-
level ethical principles, therefore, are procedural rather than
substantive in aim. Their role is not to tell us what to do in
particular cases but to provide us with standards of rele-
vance or "reasonableness" when appraisal of lower-order
rules is required. Vague and formal though they may be,
the lack of them would force the would-be critic to resort al-
together to supra-ethical special pleading controlled by noth-
ing save his own interests.
Kant's formulation of the categorical imperative has often
been criticized as an empty form from which no "duties"
could be inferred. But such a criticism, as C. D. Broad has
pointed out, based upon a complete misapprehension of
is
Kant's intention. What Kant was trying to do with what —

—
success we do not have to consider here was to characterize
the general validating procedure by which, as he believed,
all moral imperatives are ethically justified. It is not a rule
of conduct but a formula for testing rules of conduct. It had

to be "empty," it had to be formal, if it was to do the job
assigned to it. To enrich its content would be ipso facto to
transform its role and hence to deprive it of its power as
a general principle of ethical criticism. The trouble with
Kant is not that he provides us with an empty formula but that
the one he provides will not bear the burden of justification
required of it. What he saw, with an unrivaled clarity, is
that moral criticism which is something more than an ad hoc
expression of individual attitudes is impossible save on the
assumption that there are ethical principles which are gen-
eral in normative appeal.
THE POST-ETHICAL LEVEL

We come now to the final or, so I have called it, the "post-
ethical" or "human" level of moral discourse. The problem
at this levelcan perhaps best be understood in terms of the
paradoxical question "Why should I be moral?" This question
has been roughly handled by many philosophers ever since
Bradley. But in Kant it was already given, if not its quietus,
then its comeuppance. For the whole point of Kant's sharp
distinction between moral rules and counsels of prudence and
between the heteronomy of hypothetical imperatives and the
autonomy of the moral law was to render the question "Why
should I be moral?" meaningless. If on e is a rational b eing.
one recognizes that one is bound by moral laws; if one ques-
tions such laws, nothing that a rational being might say in
reply would make s ense, for by one's question one has d e- A/ * - *'
dared oneself unamenable to the rules of ethical reasoning.

To be sure, consequentialists have sometimes claimed to
justify particular moraT rules i rTTenns^bf their ena cts. But
'
this merely transfers the burden nl nbligflHon tn something

else which is taken to be intrinsically desirable on its own
account. When this point is reached, the consequentialist,
whether he knows it or not, is at the end of his rope. His
principle provides a justification of secondary moral rules
or prima facie duties, and through them particular moral im-
peratives; but the rule itself has and can have no such
justification. If the consequentialist supposes the contrary,
he is simply confused and understand the logic of
fails to
his own consequential arguments. All that he or anyone can
do in the end is to reiterate the intrinsic desirability of the
end which he believes to be the proper goal of moral action.
And if someone should stubbornly ask, "Why should I ac-
cept that as intrinsically desirable?" he has and can have no
answer. For him the question must be simply tantamount to
the question "Why should I accept as intrinsically desirable
what really is so?" which is tantamount to the tautology
"Why should I do what I really ought to do?"
The point, as we have seen, is that on its second and third

levels morality is a limited sphere of discourse with its own
and validity. One can ques-
distinctive criteria of relevance
tion the morality of an act and answer the question by
appeal to a moral rule; that rule in turn may be questioned
in relation to some more ultimate principle. But at last one
reaches a point at which no further ethical question can be
raised. The only possible way of validly answering the
question "Why should I approve x?" is to invoke some rule
which itself enunciates and delimits the sphere of moral
action.
All this must be assented to if one is to avoid confusion.
Yet I am convinced that many people have asked such ques-
tions as "Why should I be moral?" without posing senseless
philosophical puzzles and without wishing to "reduce" the
moral or the ethical to something else which it is not. And
I think that such persons are not invariably silly in asking
them, even though they perhaps have no clear notion of

precisely the kind of question they are asking or the kind
of answer they would in principle be prepared to accept.
But they would not be satisfied, nor would they be ade-
quately answered, if we should try to remove their moral
"cramps" by pointing out that the question is linguistically out
of bounds. For, although it is out of bounds on certain levels,
there is another, as I shall try to show, on which the rules
governing the usage of "ethical terms" cannot prevent it from
being raised.
In thinking about ethics, philosophers usually fail to bear
in mind the whole gamut of its dimensions, roles, and rela-
tions. In discussing the "ground" of moral obligation, they
fail to see that the metaphorical ambiguity of "ground" con-
ceals a plurality of legitimate or, at any rate, of inescapable

questions. In emphasizing the limits of moral reasoning
which govern the strictly "ethical" applications of "ought"
or "right," they forget that such limits are themselves man-
made and that the autonomy which, as social beings, we
normally grant to moral rules can itself be transcended by the
raising of questions which require the whole enterprise of
morality to justify itself before some other court of appeal.

is a many-sided proc-
Finally, they forget that "justification"
ess and that what is an adequate justification from one
point of view, is, from another standpoint, no more than the
posing of a problem. Especially is this so when we proceed
from limited questions that presuppose an institutionalized
framework of ideas and rules to unlimited ones which em-
brace the institution as a whole. If morality, like science, is
a systematic process which, at least up to a point, raises no

questions about itself, this does not mean that as human
beings we are bound to raise no questions concerning it,
even though, upon reflection, we may find that we no longer
wish to raise them. There is a sense in which man transcends
all his works and is "free," albeit at his peril, to junk any of
them at any time. I am bound by the rules of morality so long

as I am responsive to the demands of a "rational" moral
being. But nothing can give them authority over my conduct
unless I, in virtue of my attitudes and wants, am moved
by them. Moreover, no man is responsive to the rules of
morality and nothing else. It is, we are frequently told, the
office of morality to oppose and correct our inclinations.
This is true. But, if so, then it follows from the very nature
and function of morality itself that man is more if also less
than a moral being. And, as such, he may have questions
to ask of morality which it itself is unable to answer.
Nor is there anything to prevent him raising them in what
are called "ethical terms." To be sure, I cannot, or, if I would
avoid confusion, should not raise every question at once
that can be expressed in terms of "good," "right," or "ought."
But such terms have so broad a spread or spectrum of mean-
ings that there is simply no way of confining their use to
those levels we may agree to call "characteristically moral or
ethical."But there are occasions upon which I may ask, a la
Bentham, "Well, now, what's the good of all this business of
morality anyway?" And I should surely regard my answerer
as an ass if he merely replied, "I'm sorry, dear fellow, your
question doesn't make sense. Your question simply commits
a fault against usage. One cannot ask what the good of
morality is, because it is only in relation to the rules of

morality itself that you can even ask such a question."
Rather than appeal to purely supposititious linguistic rules
which prohibit the raising of the question "Why should I
be moral?" let us try to see what kind of question it is, and
in what sort of context it might conceivably be raised.
We may get a purchase on our problem by comparing
the question with others which have a similar ring: "What's
the point in living morally anyway?" "Morality is well
enough for those who choose to go in for that sort of thing,
but why, with one life to live, should I be bound by such con-
ventions?" Such questions have no single answer, nor is there
any criterion save interest itself which can determine when
an answer to the question is satisfactory. I am "satisfied" and
the question is "answered" not when some objective condi-
tions have been met but when my practical indecision or
—
doubt has been removed when, that is to say, I have been
provided with an adequate motive for playing the moral
game. Here the only sort of justification possible is of the sub-
jective sort which provides an "exciting occasion" capable of
motivating the will.
seems to me, that the existentialists,
It is at this point, it
for all their strange way

of saying things, have really under-
stood a fundamental fact of the moral life. When they speak
of man's "freedom" in the moral situation, what they mean, I
think, is that no purely logical or metaphysical "reason" can
bind a man to any obligation whatsoever, that only by a gratu-
itous decision can one in the end ever answer the question
"Why should I be moral?" I am more than my commitments. I
am bound by my commitment only so long as I continue to
be moved by it. No existential situation can compel my loyalty
unless, for whatever reason or for no reason at all, I choose
to be bound by it. Every situation, from the standpoint of
my loyalty, is a new situation, every reaffirmation of past
loyalties a new affirmation. If I choose, I may at any point
refuse to abide by past loyalties. This is not an expression
of disloyalty or of readiness to act thenceforth in an irre-
sponsible way but rather a statement of the fundamental
character of the moral situation itself. The continual possi-

bility of rejection or indifference thus renders the author-
ity of moral rules constantly dependent upon what I, as an
agent, elect to be or to do.
In the end, then, the fundamental human problem is not
to provide an answer to the question "Why should I do x?"
but "Why should I do anything?" This is a question which is
beyond reason. If it is senseless, then, as human beings, we
are at bottom committed to the posing of senseless questions.
Decision is king.
[V]
Moral Reasoning
ON THE NEW APPROACH IN MORAL PHILOSOPHY

Recently Stephen Toulmin, Stuart Hampshire, and others
have called for a radical shift of perspective in moral
philosophy. They
tell us that "traditional" approaches, which
begin with questions concerning the meanings of ethical

terms, must be abandoned. We are now enjoined to by-pass
the problem of meaning in order to reach more directly the
only really interesting philosophical issue, which concerns the
nature of moral reasoning and the use of moral judgment.
We are to ask not "What does 'good' mean?" but rather
"What sorts of reasons count as good in defending moral
decisions?" and not "How is 'ought' to be defined?" but
"What is the logic of reasoning to a conclusion that pre-
scribes what we ought The problem of definition,
to do?"
so it is anywhere, to lexicography; philoso-
said, belongs, if
phers are now to be saved from the necessity of going out of
business by the fact that lexicographers are not logicians.
Thus, apparently does Samuel Johnson leave a place in the
sun for David Hume.
Opinions well may differ in regard to the virtues of the
philosophical commitments underlying the "new" approach.
Remembering Aristotle, Hume, and Kant (to mention only
the greatest names), one may argue that it is, at best, a
second and weaker bolt from the same old blue. At any rate,
the classical moral philosophers were mainly concerned
with the problem of practical reason; nor did they waste
much time debating the question, "What is the correct analy-
Moral Reasoning 89
sis of right'?" Plato himself is only an apparent exception.

Still, the new-old approach has already helped us to break
fresh ground, or, if not this, then at any rate to sharpen the
shovels with which we dig up the old. The now ancient and
interminable debates between the naturalists, emotivists, and
intuitionists plainly reached an impenetrable impasse by
the turn of the mid-century. The only way out has seemed to
require that we go around the problem of meaning which has
stopped the "istists" dead in their tracks.
Yet it would be premature to claim that the problem of
meaning has been successfully by-passed or that the logic of
moral reasoning can be adequately explicated without regard
to themeanings of "good" and "ought." One might, of course,
question whether any very sharp boundary can be made be-
tween meaning and implication or between the sense
of a term and the conditions governing its range of applica-
tion. At bottom, meaning is a matter of our responses to
words, and our responses to words are affected by their
interconnections with other words. If the meaning of a word
is to be fully grasped only when viewed in relation to their
own wider logical contexts of argument and inference, it is

perhaps reasonable to hold that, in order to know the full
meaning of a word, one must grasp the whole syntax or
which it normally functions. Simple-
logic of the sentences in
minded questions get simple-minded answers. To the ques-
tion "What does 'good' as such mean?" no simple answer is
possible.
The trick word, here, is "fully." It does not follow that,
since full understanding of "good" requires knowledge of the
logic of normal use, no understanding of it can come
its
from knowing its definitions. Nor does it follow that, since

no sharp distinction between meaning and implication can
be drawn, there is no distinction whatever. If the full signif-
icance of a term is revealed only when its logic has been
explored, it is at least a question whether it could have a
logic at all unless it has a core meaning which remains the
same, to all practical purposes, in the different sentential
contexts in which it is used. Just as coherence theories of
go Reason and Conduct
truth, of whatever origin, are bound in theend to come a

cropper, so also coherence theories of meaning or use are
likely to lead us only to an undifferentiated ooze of larger
and larger significances. 1
It is worth bearing in mind, moreover, that preoccupation
with the problem of the definability of "good" was not due to

myopic perversity on the part of Moore and his followers.
Both Moore and his teacher Sidgwick were no less interested
in the central problem of ethical methodology than their
predecessors. They found, however, that theories of practical
reason that are detached from close study of such words as
"good" and "ought" are built upon shifting sands. Indeed, it
was precisely their desire to come to closer grips with the
methods employed in moral justification and reasoning
which forced them to consider whether, and in what terms,
"good" may be defined. The besetting sin of the then tradi-
tional approaches to the logic of morals consisted entirely in
the fact that they tended to prejudge the methodological
issue from the outset by tacitly begging crucial questions as
to the meanings of ethical terms. By challenging such ques-
tion-begging presuppositions, Sidgwick and Moore were able,
almost at a single stroke, to place in extreme jeopardy all
so-called "naturalistic theories" of the validation of moral
judgments. For unless "good" is a descriptive predicate, and
unless the property which it supposedly designates can be
identified, how can the proposal be maintained that moral
judgments are verifiable, and hence justifiable, by the
1 To say, for example, "The sky is blue" can be understood only in
so far as it is viewed as part of or in relation to something called "the
whole of science" is, in effect, to deny the possibility of knowing what
"The sky is blue" conveys to all. It is also to preclude the possibility
of knowing what logically follows from "The sky is blue." Rational
puddle-jumping presupposes islands of distinguishable meaning. If
"blue" has no distinguishable sense or meaning apart from the senten-
tial contexts in which it may be placed, then it becomes problematic
how significant assertion and denial could even occur. Possibly the rad-
ical piecemeal "meaning analyses" of the first half of the century went
too far. But one must not forget that, in part, such analyses were a
much-needed reaction against the coherence theories of the preceding
generation which, when pushed to their logical conclusions, made of
predication an incomprehensible muddle.
Moral Reasoning 91
methods of empirical science? If, in short, one is to main-

tain that "x is good"
capable of a certain type of valida-
is
tion, one must also be prepared to argue that "good" is the

sort of term whose use permits that type of validation. This
requires attention to the question of meaning itself. More
generally, it may be argued that the sorts of "reasons" that
are permissible in ethics depend upon the kinds of terms
involved in asserting that something is right or good or that
a certain act ought to be performed.

In order to determine why this may be so, let us for
a moment consider the background of the new approach it-
self. Now there can be little doubt that the present preoccu-
pation with the problem of moral reasoning is itself largely
due to the challenge which the emotive theory has offered
to the whole notion of rational justification in moral dis-
course. I dare say no one would have been upset by the
emotive theory had it not appeared to imply the impossibil-
ity of committing an error in reasoning to a moral conclusion
— if, in short, it had not seemed to entail that the entire
substance of an ethical disagreement consists in nothing

more than an opposition of first-personal attitudes, the re-
moval of which can be effected only by so-called "persuasive
methods." Not all the attacks on the emotive theory were
well founded; many seem almost perversely bent upon
misinterpreting its intention, which was meta-ethical rather
than moral. Nevertheless, beneath the confusion lay a well-
founded apprehension. For if the emotive theory be true,
not only most philosophers but most ordinary men have
been laboring for centuries under a profound misapprehen-
sion in seeking by supposedly rational means to establish
the validity of their moral judgments and decisions.
What is the nature of the challenge presented by the
emotive theory? In essence it consists almost entirely in a
radical hypothesis as to the meanings of ethical terms. In
this view, it is because ethical terms have no "cognitive
meaning" or else, in their crucial ethical role, are independ-
ently emotive that ethical arguments to a moral conclusion
are alleged to have no logical cogency whatever. The point
may be more forcibly put in another way. It has been widely

assumed that the rationality of moral discourse depends
upon the cognitive meaningfulness of its terms. Validity,
Stevenson has maintained (and many cognitivists would
agree ) applies only within the spheres of logical and factual
,
reasoning. It does not apply to mixed trains of thought,

association, and feeling, or to utterances which are merely
expressions of feelings or prods to action. Therefore, if the
fundamental meaning of ethical terms is merely expressive
and incitive, no argument to a moral conclusion can be
either valid or invalid, and one argument is as valid or in-
valid as any other.
Thus did the emotive theory of ethical terms entail, or
appear to entail, a corresponding view regardng the uses
and limits of reason in ethics. Without this the emotive
theory of ethical methods would have little to recommend
it. But the moral is clear: if the emotive theory is to be
effectively overthrown, then its claims concerning the mean-

ings of ethical terms will sooner or later have to be refuted
or else qualified. One will have to show either that such
terms have a different sort of meaning, which permits the
application of rational methods to statements containing
them, or that even if their meaning is noncognitive, it is
still such that rational processes of validation may still apply
to moral judgments. In either case one will have finally to

come to grips with the problem concerning the nature of
ethical terms.
It is no good arguing, as Toulmin does, for example, that
the emotive theory must be false since there are such things
as good and bad reasons in ethics. This will not for a mo-
ment faze proponents of the emotive theory, for they may
immediately reply that when we say that x is a good
reason, we are merely expressing our own personal ap-
proval of it and recommending that others do so as well.
Persuasive methods are plainly not limited to so-called "ob-
ject-level" statements. They may be employed on any level
of discourse whatever. The fact of emotive meaning, if such
it be, must be reckoned with not only in sentences pre-
Moral Reasoning 93
scribing what we ought to do but also in sentences pre-

scribing how we ought to argue or how we ought to weigh
statements introduced in the support of a moral conclusion.
The emotive meaning of "good" does not disappear when
we predicate it of statements or arguments rather than of
acts or objects. And when applied in the former case, or
so it may be argued, its use is no more rational than when
applied in the latter.
Nor does it much help to argue, as Toulmin also does,

that "good reason" may be translated as "valid reason."
For a skillful that, when applied
emotivist may easily reply
beyond the spheres of formal logic and induction, the term
"valid" is itself merely emotive and rhetorical. Stevenson
himself simply fell into his own trap when he sought per-
suasively to restrict "validity" to the methods of formal logic
and empirical science. The issue here concerns the meaning
of the terms predicated of an argument or process of justifi-
cation, not the mere vocabulary employed in the predica-
tion. It is not, in short, a question of whether we can say
that an ethical justification is good or valid or reasonable,
but what we mean and what we do when we say this. If, at
bottom, all we are doing when we say that such and such
an argument is valid is to put upon it our own personal
stamp of approval, then it would appear that there is
logically no way of telling another who disagrees with our
reasons that he has made a mistake. For how can one make
a mistake unless there is some rule or procedure or usage
which tells us when a mistake has occurred? The emotivist
claims that ethical arguments may be defeated; he denies
that they are in any ordinary sense corrigible. To outflank
the emotivist, therefore, one must at last come to grips
with the question of how terms such as "good" or "ought,"
the force of which is may be subject
prescriptive or incitive,
and transformation.
to rules of application
In the end, I submit, questions of meaning and questions
of logic or reason must be made to ride tandem. Neither can
be successfully treated without regard to the other. In the
next sections I shall consider certain aspects of the logic of
moral discourse; then I shall return once more to the ques-

tion of meaning.
THE MORAL USES OF DEDUCTIVE REASONING

Another feature of the new approach is its contention that
in ethics we are faced with a special sort of reasoning which
is different in kind from those employed in formal logic or in
empirical science. But here again we do well not to dismiss

too soon more traditional views.
Because ethical arguments are, at certain junctures, sub-
ject to special "moral" rules of inference, it does not follow
that, at other junctures, ordinary deductive and inductive
methods do not also apply.
Now one of the reasons why it has been argued that the
logic of moral discourse is peculiar is that it has been taken
for granted that ordinary deductive and inductive methods
are applicable only to statements which are cognitively
meaningful. Ethical judgments are, at least in part, non-
cognitive. Hence, if they are to be regarded as rationally justi-
fiable, there must be special rules of normative inference
applicable to them. This conclusion, I am now convinced,
is have no very precise notion of
also premature. Philosophers
what they mean by "cognitive meaning," and hence no clear
idea of the range df deductive and inductive procedures,
even when it is granted -inat they are to be restricted to
cognitively meaningful statements. But, apart from this,
there is no compelling reason to suppose that the applica-
tion of such procedures to normative statements or to mixed
reasonings which are partly cognitive and partly normative
is illicit. For example, we all constantly seek to justify
particular moral conclusions by subsuming them under gen-

eral moral principles. Granted that the terms of the conclusion
are, as we say, contained in those of the premises, we are
all prepared to acknowledge, in certain cases at least, that
the conclusion follows logically from the premises. All that
is required for deductive methods to apply is that the sen-
tences in question be capable of quantification and of some

Moral Reasoning 95
lawful form of negation or opposition. Sometimes the logic

required is two-valued; sometimes it may have to be
three-valued. That is as it may be. But the cogency of a
logical inference does not presuppose the "cognitive" status
of its premises or of its conclusion; it depends only upon
the question whether certain rules of transformation have
been observed. All that we have to do in performing de-
ductions in ethics is simply to carry through the appropriate
substitutions of ethical terms for the variables of the appro-
priate syntactical or logical forms.
Indeed, it may well be argued that, just because formal
logic is indifferent to questions of modes of meaning, the
only serious logical question, in the wider sense of the term,
concerns the use or meaning of such words as "good,"
"right," and "ought." Onceproblem has been resolved,
this
no special logical issue remains. If so, the only remaining
task for theoretical ethics concerns what has been called
the "epistemology of morals," that is, the process whereby
the general principles of moral discourse are themselves
validated or verified.
It is possible to reply, however, that such a view is ill
advised and question-begging, since it neglects the possibility,

brought forcibly to our attention by the emotive theory,
that the distinctively nonreferential meanings characteristic
of ethical terms preclude altogether the application of log-
ical procedures of any sort in the validation of moral con-
clusions. The logic of the books, whether two-valued or three-
valued, applies only to statements which are true, false, or
probable. Ethical judgments, however, are or contain recom-
mendations, proposals, prescriptions, and incitements to act.
These can be neither true nor false nor probable in any
intelligible sense. How, then, it may be asked, can one speak
of relations of entailment or logical implication holding be-
tween them and other expressions? The appeal to general
normative principles is not like the appeal to universal
premises from which particular conclusions are inferred. If
I accept a universal moral rule, I am bound by no law of
logic whatever to accept any particular moral conclusion. It

is true that we continually do assert moral principles in

support of particular moral conclusions. But here the relation
is again purely psychological. When we find ourselves ad-
hering to moral principles which conflict with our partic-
ular moral decisions, no question of logical inconsistency is
involved. There is merely a material opposition of attitudes
which happen to conflict with one another in action.
( To this it must be rejoined that, while there can be no
) logical opposition between individual attitudes, it does not
} follow from this that there is no logical opposition between
V sentences which express such attitudes. Again we must focus

upon the question whether, as normally used, ethical judg-
ments are related to one another in ways that accord with
the syntactical rules of logic. That we do normally so relate
them is plain. Nor does formal logic itself preclude the possi-
bility. If all s's are p, and a is an s, then a is a p regardless of
whether p is a descriptive predicate such as "yellow" or a

normative predicate such as "good." There is no rule that I
have been able to discover in the logic books which restricts
the values of the variables of any ordinary sentential func-
tion to so-called "cognitive" or descriptive terms. There is
no formulated condition that they must mean in some partic-
ular mode of meaning. If it be replied that there is at any
rate a presumption that the sentences derived by substitu-
tion from such functions must be either true or false, then it
must be rejoined that there is no good reason to deny that
ethical judgments are not properly spoken of as true or
false. Toulmin tells us that ethical judgments may be valid
or invalid but not true or false. But if his appeal is to our

ordinary moral practices, as it is, then the reply is clear; it
is far more appropriate to say that it is false that I ought to
beat my grandmother than to say that it is invalid. Toul-

min's contention is incompatible with the most obvious facts
of ordinary moral discourse. He may then wish to say that, if
we so speak, we ought not to do so. But to this the reply must
be that in saying this he changes the question and that,
when he has given us valid reasons for such a recommenda-
tion, we will be delighted to consider them.
Moral Reasoning 97
For my have discovered no sound reason whatever

part, I
for denying to our ethical judgments, even if they be ad-
mitted not to be factual descriptions or predictions, the

right to be regarded as true or false. Notice that no one
objects when it is said that a certain statement in formal
logic is Nor does anyone object to the use of
true or false.
The interesting question is thus
the phrase "logical truth."
not whether moral judgments may be true or false but
rather in what sense they are so; nor is the issue whether
this sense is identical with that intended when we hold a
statement of fact to be true or false but rather whether
there is a sufficient analogy to permit the application of
logical procedures in both cases. The application of "truth"
and "falsity" in either case signifies primarily that the state-
ments in question are validatable or certifiable in accord-
ance with certain governing rules or procedures.
Let me make the point of these remarks more explicit.
It is not denied that, in certain respects, the process of justifi-
cation in morals differs from that involved in formal logic.

It is maintained only that there is no sound reason for deny-
ing that ordinary deductive methods do apply, not merely
to the factual parts of an argument to a moral conclusion,
but also to the distinctively normative aspects. Logical sub-
sumption occurs as well in ethics or in the law as in
science or pure logic. If this requires us to believe that other
than factual statements may have logical implications or
relevancies, then we must make the best of it. And if it
requires us to adopt the view that words having noncognitive
meanings are themselves sometimes subject to proprieties,
that in the ascription of a noncognitive term we may make
significant mistakes, and that we sometimes may be required
to withdraw normative judgments on pain of inconsistency
and error, then we have our work cut out for us. For my
part I welcome the result. What is wanted is a better theory
of "noncognitive" terms.
COMPETING CLAIMS AND

THE WEIGHING OF ALTERNATIVES
But granted that formal logic has its uses in the normative as
well as in the factual aspects of moral discourse, I fear
that this does not take us very far. Our moral perplexities
are rarely removed simply by introducing general rules
under which our particular moral conclusions may be sub-
sumed. Normally the appropriate moral rule occurs to us, as
we say, intuitively, without effort or thought. In a way, all
that has been done through the process of logical subsump-
tion is to guarantee that we really are involved in a moral
situation. What usually needs further to be considered is
whether the duty laid upon us by the principle is itself to
go unchallenged. If it is merely a question of our inclina-
tion not to keep a promise, then, granted that I have made
a promise, the moral question is easily resolved. If I am
tempted to tell a whopper merely for the fun of it, then, if
there is a chance of my being believed, it is perhaps plain
that I ought not to tell it, since it is generally agreed that we
ought to tell the truth when any question of belief is at
issue. There is little room for perplexity here. If we have
trouble finding the appropriate rule, something has usually
gone wrong with our moral training. More frequently, moral
perplexities arise, not from problems of subsumption, but
rather from problems of adjudicating between competing
moral claims, each of which has its accepted justifying moral
principle or rule. The usual trouble is that our actions have
complex characteristics or effects, some of which jeopardize
or infringe upon other acknowledged claims and responsi-
bilities.
In such cases how can we rationally decide what we

ought to do? Is it here simply a question of discovering a
more general principle under which lower-order principles
may be subsumed? Does the justification of promise-keeping
or of respect for the liberty of persons depend ultimately
merely on the fact that the principles of promise-keeping
Moral Reasoning 99
and respect for the liberty of persons are specific cases or

applications of some more universal principle such, for ex-
ample, as the principle of least suffering? This would seem
to imply that there is no obligation to keep promises or to
it can be shown that
respect the liberty of others unless
promise-keeping and respect for the liberty of others is
required in order to reduce suffering. In that case, however,
we could hardly speak accurately of a conflict of duties at
all, for then we would really have only one duty which is
prescribed by the principle of least suffering. The obliga-
tion to keep promises or to tell the truth would then be no
more than an application of thismore general duty to cer-
tain specific sorts of action in which there is a presumption
that the sort of action in question usually results in less suffer-
ing than its opposite. This seems to me to be highly dubi-
ous. Let us see why.
Suppose that there is a question whether we should keep
a promise which we have made in good faith. Two consider-
ations need to be observed. In the first place, the very mak-
ing of a promise may reasonably be said to carry with it
an obligation to keep it. Other things remaining equal, any
promise creates an obligation which cannot be denied with-
out committing a moral fault. But other things are not al-
ways equal, and in certain situations it may be that I ought
not to fulfill my obligation to keep my promises. No one
would deny that, although we ought normally to keep our
promises, other claims may be introduced which may chal-
lenge the obligation to keep a promise. Suppose, for example,
that the net effects of keeping my promise will most prob-
ably result in a greater suffering than the effects of breaking
it. Then it may be argued that a genuine perplexity has
arisen, which in effect requires us to choose between com-
peting principles of conduct. Now it may be held that unless
the suffering entailed by keeping a promise were substan-
tially greater than that entailed by breaking it, the promise
should be kept. The principle of least suffering does not
automatically take precedence over the principle of prom-
ises. It does so only at a certain point, when a great deal of
ioo Reason and Conduct
suffering would result from the keeping of a promise. It is

only in this sense that the principle of least suffering takes
precedence over the other. But how much suffering is re-
quired to make the principle of least suffering take prece-
dence? Here, I think, there is no exact prescription and no
clear way of guaranteeing that, granted equal knowledge of
the facts of the case, reasonable men may not disagree
about the point at which the deleterious effects of keeping a
promise absolve one from keeping it.
Nevertheless, I believe that most interested persons in our
society would agree that in extreme cases the principle of
least suffering does take precedence over that involved in
making promises. But let us consider a more serious per-
plexity. Suppose there a question whether we should
arises
perform an act which, as we believe, will result in less net
suffering for the people concerned than any feasible alterna-
tive but which cannot be performed without serious injustice,
that is to say, without piling up unequal misery upon a
single person. Now the ordinary utilitarian would have to
hold here that, our only obligation being to maximize the
happiness or minimize the misery of those who are in any
way affected by our conduct, the obligation not to act un-
justly is derivable solely from that to maximize happiness or
reduce misery. There are also some philosophers who would
admit that the demands of justice impose independent obli-
gations but who would also argue that these obligations must
give way when they seriously conflict with the demands of
the principle of least suffering.
To the ordinary utilitarian we must reply that most of us
raise no question about our obligations to be just in our
dealings with others unless, in a particular case, it is alleged
that by fulfilling the obligation a greater moral evil would
result than by violating it. We ask for a reason when someone
tells us that we should act unjustly; we do not ask, other
things apart, why we should be just. The principles of
justice are binding upon us in their own right. They belong
to the fundamental commitments of our moral universe,
as the events of recent years have forcibly brought home to
Moral Reasoning 101
us. To ask us to regard them otherwise would be tantamount

to asking us to give up some of the very justifying principles
in terms of which we conduct our deliberations. There is no
"because clause" attached to the principles of justice, for
they are among the principles in terms of which reasons are
given ur trying to justify our actions. To ask why one should
be just istantamount to asking why one should be moral.
To this no answer can be given within the framework of
moral discourse as we know it. The utilitarian, in effect,
simply asks îs to renounce what, for us, is the moral life
itself.
But what shall we say to the moralist who maintains that,

though the principles of justice may be independently
binding, they must give way when they require the per-
formance of actions which conflict with the principle of hap-
piness or least suffering? Here, I think, we may ask for a
reason. One cannot just appeal to the principle of least
suffering itself, since it has been shown that the principles of
justice do not derive their moral force entirely from the
former. Is it, then, a matter of self-evidence? This is easily
shown not to be the case. Suppose that an act of great
injustice would result in a very slight reduction in suffering
to the greatest number of persons involved. In that case
surely we would hold that the principle of justice takes
precedence and that it would be wrong to perform the un-
just act. And in cases where the injustice is as great as the
utilitarian advantage from acting unjustly, most of us would
say that there is no clear answer to the question of which
is morally preferable.
It is my contention that, although in many cases it is
normally possible to decide which, among competing claims,

is to be accepted, in at least three sorts of cases there is no
conclusive reason for always giving one claim the right of
way over the rest. I shall call these the claims of least
suffering or humanity, the claims of justice or fair treatment,
and the claims of liberty of persons. 2
2 Consider, for example, the case of Skinner's Walden Two. Suppose

you are a very kind and extremely knowledgeable person who is in a
Now it may be argued that no reason worthy of the name

can accept such a situation and that beyond any commit-
ments of the sorts I have mentioned there is still another
and more fundamental one which enjoins us, whenever we
find ourselves involved in a conflict of duties, to search for
a further principle or procedure that will enable us to ad-
judicate decisively between them. It has been said that the
supreme principle of rationality is the commitment to be
consistent in thought and action, so that, if we find that our
moral code involves us in inconsistency, the cade is so far
irrational and must be modified. However we act, whether in
the name of least suffering, or justice, or liberty, or what-
we
are bound as rational agents to choose as our prin-
Iever,
cipleone which would be equally binding upon all rational
agents in a similar situation. Either one is committed, then,
to choose once and for all one of the preceding claims as
prior, and to hold it as unconditionally and equally binding
upon all rational beings, or else one is committed to search
for some still higher principle which takes precedence over
position to institute, at your pleasure, a society of happy, carefree men,

on the sole condition that you deprive them of all liberties of person.
This means, in part, that they will have no choice with respect to their
vocations or their mates; that they will not have the right to criticize
the policies or theories of their rulers; and that there will be, in short,
no important area of human activity in which they will be free from the
benevolent supervision and control of the ruler. Privacy and freedom
from benevolent interference in such a society will have no place. We
must, of course, play the game fairly. We will suppose, therefore, that
our happy people have been given an elixir so that they will not know
or, if they know, then not resent what they have lost; that, in short,
they will no longer care for privacy or for the freedom of their persons
at all. On the other hand, although they will be manipulated by the
ruler, it will always be for their own good, so that they themselves will
never dream of making a complaint that the liberties which they do
not enjoy have been violated. It is my contention that, although such a
choice, if presented to us as possible rulers, would have something to
commend it, we would still not be able to say that a sufficient reason
has been given for adopting it. For I think we would maintain that the
freedom of persons ought categorically to be respected and that, when
they seriously conflict, the principles of least suffering and happiness do
not always override the demands of freedom. What are at stake here
are the basic commitments of a whole way of life, neither of which is
finally and unconditionally prior to the other.
Moral Reasoning 103
any of these and, through it, to reach a decision which

would also be so binding.
In this form, the argument is pretty clearly mistaken. For
in the first place, although it is true that, as rational agents,
we are bound to try, as far as possible, to be consistent
in our actions, and that we should never rest in a conflict of
principles without first carefully searching for a more ulti-
mate principle,can scarcely be held that we are bound to

it
discover such a principle or that we may not in conscience

find that among our
ultimate commitments there will be some
that, in practice, do seriously conflict. After centuries of
search, moral philosophers have found no one supreme prin-
ciple which in hard cases all reasonable men must in con-
science acknowledge as prior in its claims to every other. On
the contrary, I believe that, if we are candid, we will have
to admit that we are morally certain that there are equally
binding principles which, at least in hard cases, do involve
us in fundamental conflicts of duties. It cannot reasonably be
asked that I renounce the very principles in terms of which
the whole process of moral justification is carried on or that
I must choose between them once and for all when they
appear co-equal. How shall I do so, and on what grounds?/
It is nothing to the point to argue that another sort of being
might find himself involved in no such conflict of principles
or that in Paradise thedemands of freedom or justice might
be held to be secondary or even practically meaningless.
We do not inhabit Paradise. The only sort of practical reason
which is relevant for us to consider is that which involves
the procedures of justification to which we ourselves are
committed and in terms of which alone we give reasons for
what we ought to do. In one sense, ultimate moral principles
are, as some philosophers say, given; it may be that to
certain other creatures they would not be given and that
for them other procedures of justification might appear quite
as proper or sensible as ours to ourselves. But it is our
business, as moral philosophers, to uncover if we can the
principles which are given and which define
to ourselves
what we ourselves understand by sound or unsound reasons
in moral deliberation. It is our own moral consciousness that

concerns us here. If someone wishes to move that this
be changed, he may; that is another matter. But his recom-
mendation will not of itself carry any rational obligation on
our part to accept it. It does not matter to us that in 1984
no one be bothered by the demands of justice or that in
will
Walden Two no one will miss his lost liberty; that is not our
affair. It is our own procedures of justification or of practical
reason that we have to analyze. With respect to them, I
maintain, fundamental conflicts of principle are clearly pos-
sible.
In the second place, I submit that in our system it is

assumed that in certain situations, and with all evidence be-
fore them, men may still reasonably disagree and that, upon
reflection, they may properly decide differently as to the
weights to be attached to the respective claims of liberty,
justice, or least suffering without fear of justifiable recrim-
ination or blame on the part of their peers. But, if this is so,
then it cannot be maintained that, as a rational moral agent,

I must always decide precisely in such a way as all other
men would decide in a similar situation or even as I myself
would subsequently decide in the same sort of circum-
stances. Upon reflection I may decide to weigh the con-
flicting claims differently without involving myself in the
charge that either in one case or in the other I have acted
unreasonably. Formulas here are of no help.
Nevertheless, it does seem that something further may be
said about the demands of reason in such a situation. For

itmay be said that, when I am involved in a conflict of basic
commitments, must decide between them and that I can-
I
not justifiably renounce them both. Moreover, it is always in

order to demand that a man reconsider and that he review
both the facts and the principles in question in order to
make certain that no vital point has been missed. In short,
I am always bound to listen to reason, and this means, in
part, that the process of moral reconsideration and justifi-
cation is never finally closed, even when, for the time being,
I find that I must provisionally make up my mind. But, having
Moral Reasoning 105
reconsidered, then as a rational agent I am free to choose

for myself between the conflicting principles, and, in so
doing, I may claim to have done everything a reasonable
man may be required to do in order to reach an impartial
and objective decision. And my decision in such a case will
be impartial and objective, since no one may fairly claim
that I have failed to do what any man ought to have done
in the same circumstances, even though some men might
reasonably have decided differently.
What emerges, here, is that there is finally a direct appeal
to "reason" itself rather than to the specific principles that
bear its name. 3 I am bound to be just; but I am also bound
to listen to reason when justice conflicts or appears to con-
flict with welfare. This does not commit me to decide against
justice, but it does commit me to consider the possibility
that Imay have been mistaken and that the weight should
perhaps go the other way. This means not that we are here
appealing to some special faculty of moral intuition but that
we are invoking still another regulative procedure which
claims governance over the process of deliberation itself.
It does not tell us in particular what our first-level duties

may be; it provides no formula for weighing the respective
claims of conflicting obligations. It tells us only that we
must take an impartial or general view, that we must con-
sider how others in our predicament have acted, and that we
be prepared to review the facts of the case upon demand.
It tells us, in short, what steps we must have taken in our
process of justification if our choice is to be held free from

blame. This is a procedure to be used in trying to make up
our minds about claims, but it is not itself an injunction to
make up our minds in a specific way. Like Kant's supreme
principle of categorical imperatives, it prescribes the form
which the deliberation prior to a justifiable decision must
take, not what the decision itself must be.
3 Cf essay VII, "Global Conventions
. in Ethics," pp. 128-33.
io6 Reason and Conduct
THE MEANINGS OF ETHICAL TERMS AGAIN

It was my contention earlier in this essay that questions of
meaning and questions of reason in ethics cannot be com-

pletely separated and that an adequate theory of reason-
giving involves a theory regarding the nature of ethical terms.
I wish now
go a step further in trying to make good this
to
claim. In so doing, Ihope to shed a further light upon the
character of the deliberative processes involved in moral
justification.
On all levels ethical terms have a normative aspect which
cannot be adequately explicated in terms of their contextual
descriptive meanings. 4 This is true also of the processes of
justification in ethics; they too are normative in intention
and in use. Such terms as "valid" and "invalid," "relevant" and
"irrelevant," "ethical" or "unethical," and, finally, "rational"
and They are used to
"irrational" are themselves normative.
commend condemn supporting arguments and, indirectly
or
through them, the choices or decisions which may depend
upon them. To some this may suggest that after all the
present view differs in no important respect from the
emotive theory and that all the preceding talk about princi-
ples and reasons was simply an elaborate way of commend-
ing certain ways of resolving moral perplexities, without
any theoretical significance.
I think that this is not true, and I shall now explain
briefly wherein the present view differs from ordinary emo-
tive theories like that of Stevenson. Let me say at the outset,
however, that do not wish to deny that we often do use
I
words like "good" or even "ought" in order to express our

first-personal sentiments or to issue first-personal directives
regarding which there is no thought of rational justification.
If I should say "Good play!" at a football game, I would
4
That ethical terms are used descriptively, as well as normatively,
isa thesis which underlies all of my work in ethics and value theory.
This thesis, if correct, precludes the need for serious controversy be-
tween so-called "cognitivists" and "noncognitivists."
Moral Reasoning 107
perhaps merely be expressing my own emotions and inciting

others to similarly vent theirs. And when I reply, "Right,"
when asked if I plan to be at a certain place at a certain
time, I may be simply expressing my own intention to be
there. But there are other occasions when the ordinary emo-
tivist be plainly
characterization of these predicates seems to
inadequate. For, in the first place, there are, as I have in-
dicated, systematic connections between sentences contain-
ing these terms which are by no means arbitrary. Some rea-
sons, regardless of how we may personally feel about them,
are generally accorded a relevance which is quite independ-
ent of their capacity to win an argument for us in the purely
rhetorical sense. In speaking of a reason as irrelevant, we do
not mean simply that we ourselves wish it to be discounted;
we are condemning from the standpoint of a system of
it
be equally binding upon

justifying procedures that are felt to
all members of our moral community. We speak of certain
reasons as valid or invalid and of certain moral decisions as

reasonable or unreasonable. When we do, we are asserting
claims —not in our own persons, but rather in the name
of a set of principles to which
normal persons in our moral
all
community are committed. In short, the moral judge or critic

acts as the voice of an impersonal system of prescriptions
and procedures which are impersonally regulative of our
deliberations.
To speak of a reason as "valid" is, then, to commend it.
But it is not just to commend it; it is, rather, to commend it
in a certain way and in accord with a general method of
commendation. What needs to be remembered is that there
may be impersonal conventions of praising and blaming just
as there are impersonal methods of validating logical or
factual truths. There is a vast difference, not merely in tone,
but in meaning or use, between "Please pass the butter"
and "You ought to keep your promise," or between "Fine
weather we're having" and "Every human being ought to be
at liberty." The difference lies precisely in the fact that
moral recommendations are subject to a social ritual of im-
personal justification or authorization which limits the way
in which supporting reasons are introduced and corrects us

when we make a mistake. In short, there are normally tacit
"because clauses" and "unless clauses" written into particular
moral prescriptions which cannot be correctly or properly
filled out in any way we please. And what is meant here by
speaking of "properly" filling out such a clause differs in no

important way from what is meant when we speak of
errors or improprieties in any other sphere. Terms such as
"proper" are, if you like, evocative and persuasive. But they
are no more mere first-personal evocations of feeling when
used by the moralist than when used by the logician or by
the grammarian.
When the question "Why?" is raised, it functions, so to
say, as a signal which shifts the context of the use of ethical
terms to a systematic plane where the concepts of validity
and relevance may begin to gain a purchase. At this level it
becomes legitimate to ask for bona fide reasons and to
speak of reasons which meet or fail to meet certain condi-
tions as "sound" or "unsound." In so doing, we have not
passed beyond the normative sphere. We have merely in-
troduced a framework of systematically related procedures
of justification by which, as we say, rational persons within
our community are governed. Such procedures are them-
selves expressive; but what they express are not just the
first-personal sentiments of the person who invokes them but
rather, the impersonal social presses of the whole community
of law-abiding men. This is why their "weight" upon us is so
ponderous and why we are so loath to violate them.
In partial justification of this view, we may consider how
we normally interpret the norms of morality when viewed
as natural signs. What, in fact, do we normally take them to
signify? I submit that no one regards them merely as signs
of the sentiments of the speaker. Rather do we construe
them as indicative of certain general social attitudes or sen-
timents which we refer to by such expressions as the "con-
science of the community" or even perhaps as the "con-
science of mankind." The moral critic, in using the language
of reason, speaks for and in his person represents the pre-
Moral Reasoning 109
sumed collective sentiments of a group. On its prescriptive

side, moreover, the moral judgment is not an isolated or
blind evocation of passion or feeling toward an object. The
presence of the because clause, and the admission of the
relevance of "Why?" indicates that the incitive role of "ought"
is qualified —that it carries with it a train of normative con-
ditions and implications which cannot be turned on or off at
will.Moreover, the ethical "ought" appeals to us as men and

not as individuals. It addresses us, not as particular John
Smiths or Bill Joneses, but anonymously, as rational men
who are governed by the procedures of an impersonal moral
order. This framework of rules and principles, enables us to
speak of the misuse of "ought" or of the falsity or invalidity
of a particular "moral judgment." In virtue of it we carry on
a regularized process of correction when errors in judgment
or in reasoning are made. In a word, it is possible to make
a mistake in morals.
Second-level words of moral discourse, such as "valid,"
place a stamp of collective approval upon certain modes of
deliberation and justification. They prescribe, as the case
may be, what is socially permissible, what is mandatory,
and what is prohibited in the matter of forming our decisions.
He who uses such terms is at once caught up in a process of
regularized reason-giving which has nothing to do whatever
with his personal preferences or wishes in the matter.
I do not wish to claim too much. I have not said that the
moral rules of rational men cannot be altered; nor have I

held that the procedures for justifying first-order moral de-
cisions yield sufficient reasons in every case. Quite the con-
trary. There can be no question here of a calculus. Even
to speak of decision procedures would be misleading if it
suggested that we possess well-formed rules for the weighing
of obligations in cases of conflict. There is perhaps a greater
openness of texture in moral reasoning than in ordinary fac-
tual reasoning. This does not imply, however, that in morals
anything goes so long as it is concept
effective or that the
of validity has no legitimate application moral delibera-
in
tion. Reasonable men may reasonably disagree, but there
no Reason and Conduct
are still many alternatives which all of them acknowledge
to be arbitrary and irrational. If that is so, then any simple

emotive theory is surely mistaken, and this on its own
grounds.
[VI]
The Authority
of Moral Judgments
PREVAILING ANALYTICAL MODELS

IN ETHICAL THEORY
Until recently, most philosophers have taken the statement

of fact as their model in analyzing the meaning of ethical
judgments. This is true not only of the naturalists, who
have sought reduce ethical judgments to empirically veri-
to
fiable statements, but also of the non-naturalists, who deny
that such a reduction is possible. For both groups questions of
fact are posed as the primary concern of ethical controversy,
and the authority of any ethical judgment is accordingly
assumed to be directly proportional to the validity of its
factual truth-claim. When

any question is raised concerning
one's commitment to do what the moral judgment asserts
or demands that one ought to do, the descriptivists for —
—
so I shall call them here usually brush it aside as a "psycho-
logical question" which is therefore irrelevant to the problems
of ethics. To be sure, some descriptivists such as Dewey have
emphasized the decisional or practical context of moral dis-
course; but its de jure or normative status is generally
treated as a simple function of its claim to truly describe the
objective characteristics of actions conceived as processes or
as events.
However, if by the "normative" status of a judgment is
meant its role as a guide to conduct, then it is clear that
the normative status of ethical judgments cannot be ade-

quately understood in terms of a descriptive model whose
primary function is to convey factual information. For the
relevance of any description of fact to the principles which
control our actions is not logical, but motivational; or, put
in another way, it is not what a statement descriptively as-
serts which establishes its authority as a guide to action, but

its power, by whatever means, to influence the will. It is for
this reason, I believe, that the emotive theory of ethics has
had such a vogue in recent years. For the emotive theory,
whatever its merits in other respects, is designed explicitly
to account for the practical or normative aspects of moral
discourse.
Now the proponents of the emotive theory, like the non-
naturalists, usually begin by insisting that ethical judgments
in their characteristic use are not empirically verifiable, or,
what comes to the same thing, that ethical terms cannot
be defined through other terms which stand for natural or
empirical properties. But here the resemblance ceases. For
the non-naturalists, still clinging to the model of the factual
statement, are compelled to invent a realm of non-natural
facts or quasi-facts in order to give countenance to the
ostensible truth-claim of the moral judgment, while the emo-
tivist,proceeding on the basis of a completely different
model, is free to draw an entirely different concluson from
the same assumption.
The earlier versions of the emotive theory, however, were
formulated not by specialists directly concerned to charac-
terize the peculiar functions of moral discourse, but by
methodologists of science, such as Rudolf Carnap, who were
primarily interested in the problems of descriptive meaning
and truth, and by semanticists, such as I. A. Richards, whose
approach to the non-descriptive aspects of language was
through the study of literature. For the former group the
category of "expression" provided a convenient semantical
wastebasket into which could be dumped all forms of dis-
— —
course metaphysical, poetic, or moral that failed to meet
the descriptive canons of empirical verifiability. For the
The Authority of Moral Judgments 113
latter group, the primary interest in non-descriptive dis-

course was in its immediate emotional effect. And for both,
therefore, the personal utterance or evocation of feeling
became the principal model for the interpretation of ethical
judgments.
Gradually, however, it became apparent that such a model,
although stimulating and suggestive, failed to do justice to
the specific differentiae of moral judgments even when con-
ceived as a sub-class of "expressive judgments." In the first
place, as John Dewey and others insisted, the relation of

"venting" or "expression" which holds between a statement
and a speaker sheds no light whatever on the distinctively
practical intention and effect of moral persuasion. What is
wanted, in short, is not a theory of the causes of the moral
judgment but rather an analysis of its conventional use as
a guide to conduct. And from this standpoint, a linguistic
venting of emotion is not more but less relevant than a
trusted statement of fact which at least directs our atten-
tion to something to which we actually do aspire.
This point is, I believe, a pertinent criticism of those ver-
sions of the emotive theory which characterize the non-
descriptive aspects of the moral judgment simply in terms of
personal emotional expression. More particularly, it is a
Carnap, Ayer, and even of Stevenson's "work-
fair criticism of
impugn the importance of
ing models." It does not, however,
the theory in bringing forcibly to the attention of ethical
theorists the urgent necessity of going beyond purely de-
scriptivist analyses of ethical terms and judgments, if they
hope to account for the normative functions of the latter.
Moreover, it is only fair to say that some of the emotivists

— and this is particularly true of Stevenson —never sought to
construe ethical judgments exclusively in terms of the ex-
pressive model. From the first Stevenson has insisted upon
the "magnetism" of ethical terms, 1 and his whole analysis of
persuasive definitions assumes that such terms have mean-
1 See Stevenson, "The Emotive Meanings of Ethical Terms," Mind,

1947, pp. 14-37.
ii4 Reason and Conduct
ings which are independently incitive as well as expressive

in character.
Also Karl Popper, despite his radical separation of factual
from ethical assertions, regards the latter not as mere ex-
pressions of feeling but as "statements of norms," which have,
according to him, an objective thrust which is analogous, al-
though never reducible, to the statement of a fact.
It is significant, however, that even for Stevenson and
Popper, the closest analogue of the moral judgment, in its
active function, is the decision or command whose force or
authority derives primarily from the interests of the partici-
pants in a moral colloquy. This is evident, in the case of Pop-
per, from emphasis upon the fact that we are
his reiterated
always "free" to choose our norms as we please, and that the
authority of the norm derives from our avowal of it, rather
than vice versa. It is apparent also, in the case of Steven-
son, in his tendency to characterize an exchange of moral
judgments as the expression of two opposing commands
whose incitive appeal seems, at least, to be identified with the
fact that the respective speakers have expressed them.
NEW WORKING MODELS IN

RECENT ETHICAL THEORY
Thus, the characteristic model in terms of which more recent
proponents of the emotive theory have conceived the moral
judgment is essentially that of the individual proposal, de-
cision, command. But it is precisely the adequacy of
or
even this model to do justice to the full significance of moral
judgments on their normative side which has recently
—
been challenged by two philosophers Margaret MacDonald
—
and C. E. M. Joad who otherwise have next to nothing
in common. Indeed, their agreement on this point is all the
more striking precisely because they have such diametri-
cally opposed axes to grind, and because their own respective
ways of accounting for the normative significance of moral
judgments are so utterly different.
Mr. Joad's opposition to the emotive theory seems to be
based as much upon moral as upon theoretical grounds. And

I am not concerned with the former.
2
What interests me is
his contention that "ethical judgments are recognized as
claiming an authority and a publicity that feeling judgments
do not. We expect other people to share our ethical
judgments and feel that they are morally obtuse if they do
not, and we expect ourselves and others to act in accordance
with their dictates and feel that we and they are wrong if
we do not." 3 If this statement is even roughly true, and I
think that most persons would tend to agree intuitively
that it is, then no ethical theory that regards ethical judg-
ments as merely expressions of personal decisions or as in-
dividual incitements of attitudes can possibly be regarded
as providing an adequate general analysis of the normative
functions of ethical judgments. Whether we like it or not, an
impersonal, public authoritativeness is frequently claimed
for and perhaps voiced in moral judgments on certain levels
that is independent of and, indeed, precisely opposed to
the private inclinations or preferences of either the person
judging or the person judged. And this poses a problem of
interpretation, therefore, for which the ethical theorist must
provide some answer.
2 There are the all too familiar charges that to deny the objective
truth-claim of moral judgments is, in effect, to open the floodgates of
unreason and moral irresponsibility, to encourage cynicism in others,
and to irrevocably weaken the hold of one's own moral commitments.
Most of this, of course, in the sense in which Joad means it, is itself
a form of irresponsible nonsense. It cannot too often or too forcibly be
expressed that such writers as Carnap, Stevenson, and Popper have no
intention of belittling the importance of rationality in the conduct of
life. On the contrary, and this is particularly true of Popper, the whole
polemic is directed to the defense of reason in conduct against those
misguided crypto-rationalists who identify "goodness" and "obligation"
with some trans-empirical or non-natural quality which can be discerned
only by some supra-rational intuition. Indeed, their writings should pro-
vide a source of comfort to those who still retain some faith in the princi-
ples of Enlightenment, and would have the deliberations of men directed
toward the probable means to their own well-being rather than clouded
or swayed by the power of words disguised as the will of God or as the
voice of some transcendental "moral order of the universe."
3 C. E. M. Joad, A Critique
of Logical Positivism (Chicago: The
University of Chicago Press, 1950), p. 139.
n6 Reason and Conduct
The person voicing a moral judgment is, or is usually re-

garded as, merely the spokesman of morality; and the per-
son judged is singled out for reprobation primarily as the
violator of a rule that applies not merely to him but to all
members of a certain class or group. But if this is so, then
the full prescriptive meaning of "This act ought not to be
done" cannot be adequately represented in terms of the
usual emotional schema which characterizes it as "I approve
of this; please do so also."
Now it may well be true that we rarely do utter moral
judgments unless our own feelings partially coincide with
the attitudes which they prescribe. And as a rule the incitive
appeal of such judgments has and is expected to have a
bearing upon the conduct of the particular persons to whom
we address them. This is not here in question. What is ques-
tioned is that the kind of authority expressed in and exerted
by the moral judgment is adequately characterized in terms
of the emotional dispositions of the particular individuals en-
gaged moral discussion. On the contrary, just as what we
in a
call "the"meaning of a word in ordinary language is an in-
terpersonal rule which thereby functions prescriptively for
those whouse the word, so also the moral authority of an
ethical judgment is primarily due to the fact that it is
validatable by a general rule of conduct which is binding
upon the individual person only because it is regarded as
binding upon all moral persons.
THE CEREMONIAL USE OF LANGUAGE IN ETHICS

It is at thispoint that it becomes relevant to consider Mar-
garet MacDonald's important recent essay, "Ethics and
the Ceremonial Use of Language." Miss MacDonald offers
an explanation of precisely those normative aspects of moral
judgments which are not accounted for by the usual ver-
sions of the emotive theory, without, however, resorting
to the questionable metaphysics of Mr. Joad. In what fol-
lows I shall make use of parts of Miss MacDonald's analysis
in my own way, qualifying it or amplifying it as I see fit.
Now according to Miss MacDonald there are five general

characteristics which hold of all moral judgments. Such
judgments she maintains, are all "certainly ( 1 ) normative,
(2) authoritative, (3) public, (4) indicative in grammatical
form, (5) practical." 4
It is Miss MacDonald's primary contention that moral
judgments derive these peculiar characteristics partly from

the fact that they are used in a way which is akin to, al-
though not identical with, the use of language in the per-
formance and facilitation of common tasks or actions. In this
use, language may also have a "cognitive" or descriptive
meaning, but its function is not primarily to say something
about a proposed act nor to express or incite feelings, but
rather t o^ exp edite action. Similarly moral judgments are
neither mere speculations as to what might be a good thing
to do under such and such conditions nor mere initiatory
incitements to feeling; they are, rather, integral parts of
larger processes of action which we designate by the term
5
"conduct."
But not just any common task or regularized action is a
form of "conduct" in the moral sense. And not just any active
or performatory use of language is thereby ethical. We have
still the problem, therefore, of finding the specific charac-
teristics which differentiate moral discourse from other forms
of performatory utterance. To solve this problem, Miss Mac-
Donald, following certain clues of the anthropologist Mali-
nowski, proposes the analogy (which she is careful to insist
is merely an analogy) of the ceremonial and ritualistic
uses of language. And the suggestion is a happy one, for as
Malinowski describes it, ceremonial language does have pre-
cisely those traits of publicity, normativeness, and author-
ity which are characteristic of so much moral discourse.
Miss MacDonald has neglected, however, one vital part
6
of Malinowski's brilliant analysis of primitive rite and ritual.
4 Cf. Philosophical Analysis, p. 212. A collection of Essays, edited by

Max Black (Ithaca, N.Y.: Cornell University Press, 1950), pp. 211-29.
5 This last point is my own.
6 See Bronislaw Malinowski, for example, Magic, Science and Re-
n8 Reason and Conduct
As he conclusively shows, no ritual can give significance to an

act which is without an independent interest or vital func-
tion of its own. In primitive societies, the death ritual, the
marriage ceremony, the rites of hunting and of war, are all
rooted in activities or events which have, quite apart
from their ceremonial emphasis, a profound social and psy-
chological importance to the persons involved. The ritual, in
short, does not so much create the value as to "frame"
and solemnize it. In part, it serves notice, as it were, that the
acts are to be approached in a special mood and performed
with special seriousness. Yet it is never more than an aspect
of some wider activity which has a significance of its own,
without which the ceremonial language would be an empty
shell, without interest or power to move.
But what, more precisely, does the ritual do to the acts

which it solemnizes? To say that it sets apart and deepens
their significance is only partially true. For as in the case of
the death ritual, the effect is not to augment the sorrow and
fear of those who perform it, but to transmute and alleviate
it. But whatever the effect in the particular case, the ritual
invariably provides a social vehicle by means of which the

interests of the individual are emotionally tied to or iden-
tified with those of the group to which he belongs. In-
variably the effective ritual is traditional, formal, and im-
personal. Through it the individual joy or sorrow, love or

hate, is regularized and ordered in such a way as to trans-
form it into a social gesture which reaffirms the individual's
solidaritywith the tribe or clan and vice versa.
For our present purpose, then, the important thing is that
the ceremonial provides an authoritative social sanction to
acts which are already in their own right important both to
the individual person and to the group. By implicitly under-
lining the communal importance of the ritual act, the sym-
bols help to effect a unification of personal interests within a
framework of common customs and rules. Thus they achieve
ligion (Garden City, N.Y.: Doubleday, 1954), pp. 47 ff.; see also his
Crime and Custom in a Savage Society (New York: Harcourt, Brace,
1926).
their effective authority both from the private sentiments

which the separate individuals bring to the ceremony and
from the traditional social and interpersonal context into
which the ceremony automatically places the individual. The
meaning of the ritual is reinforced and, hence, subsequently
revitalized by the special concern or emotion which the in-
dividual brings to its performance; but it is never simply an
expression of personal feeling nor a vehicle of personal ex-
hortation or supplication. It always has the effect of a social
"rule" the authority of which is "in" the rule itself, but only as
a symbolic surrogate of the collective interests which it repre-
sents.
Coming to still closer grips with the problem, Miss Mac-
Donald next introduces the model of law which itself is an
instance of the ceremonial use of language. The legal ver-
dict, as she says, is not merely a statement of fact, although
it may "contain" such a statement, and although it emerges
from and upon evidence and reasoning. 7 It acquires
"rests"
its special status and authority, not primarily from the
cogency of the arguments which have occasioned it, but

rather from its ritualistic setting. The law thus does not
describe, it imposes; it not merely announces the opinions of
the jury or the attitude of the judge, it pronounces through
them upon the individual in the name of society, or more
figuratively, it is society in any visible and articulate sense.
One may of course always "question" the authority of a
law. But what does this mean? It means primarily that the
spell of the ceremony of the law has, for the nonce, been
broken, that it has run counter to interests which are so
exigent that the prestige of the law for the moment is im-
pugned. In short, to question the authority of a law is, in
part, to oppose to the sanctions of public reprobation the
more urgent motives of personal need. Merely to express
7 It is worth pointing out in passing that legal "evidence" and legal
"reasoning" are themselves shot through with ritual. Hence even the
"supports" upon which a verdict is based are more than mere facts and
deductions. Their proper weight is a function of the way in which they
are introduced. Thus, for example, facts which are not "sworn to" on
oath have no status as evidence which can be used against a defendant.
one's need, however, is socially ineffectual. As a rule, there-

fore, such a question is stated, not as a bald expression of
individual preference but, rather, as an impersonal appeal to
some higher legal authority. And this authority in turn is ef-
fectively derived from a further ceremony. Miss MacDonald
concludes from this that "Even in an authoritarian state,
the Leader must distinguish, by some formalities, laws
from other personal decisions. So the authority of law itself
is inexplicable without the ceremonies conducted by their
characteristic modes of utterance, which attend its intro-

8
duction."
Following these ceremonial analogies, we may perhaps
begin to understand the peculiar impersonal social authority
and publicity of moral judgments without inventing a super-
sensory realm of "values" to account for the fact. We may
also see why they are usually stated in the indicative rather
than in some other form which would distinguish the norm
as emanating from a personal source alone. "Do this" in
many contexts frequently provokes a counter response which
defeats its own purpose. But "It is wrong to do this" re-
moves the arbitrary decisional aspect of the explicit per-
sonal command or imperative. So understood, the moral
judgment derives its authority neither from the emo-
tional tone of the speaker nor from his physical presence
alone. As such it appeals not to the private inclinations of
the listener but to the socialized dispositions which he
possesses as a member of society. The speaker pronounces
the judgment, as it were, not as an agent, but as a carrier
of meanings whose determined elsewhere.
effective appeal is
And the listener is moved by it hap-

it, not simply because
pens to coincide with some half-aroused private emotion
which he happens already to feel, but because he is a so-
cially conditioned organism, and the rules of society are
already written into his nervous system as conditioned pat-
terns of response.
There is, however, one important hitch. And to this Miss
MacDonald herself provides the principal hint. In com-
8 Op. cit., p. 226.
meriting upon the limitations of the emotive theory she

says:
The emotive theory dissolves the hardness of moral judgments

into the softness of a romantic preoccupation with a personal
gospel and a private missionary society. It is a lucky accident
that our gospels sometimes agreeand that intercommunication
occurs between the one-man sects. This seems an exaggerated
moral protestantism. 9
Precisely. But I wonder if the implications of this passage

are not somewhat different from those Miss MacDonald has
drawn from it.
THE LIMITATIONS OF THE CEREMONIAL THEORY

Let me
begin by pointing out that the ceremonial theory
may be regarded as a species of emotive theory. It
itself
does not, to be sure, interpret the moral term along the

lines of Stevenson's first pattern analysis. But as Stevenson
himself emphasizes again and again, his first pattern schema
does not fit all of the inflections of the moral judgment. In
certain contexts "x is good" may be perhaps most appro-
priately characterized as "I approve of this; do so as well."
But it may alsomean "we approve of this; do so as well," or
"it is approved, do so as well." The important thing, in any
case, is not the expressive side of the emotivist translation,
but rather, its "incitive" aspect. This incitive meaning may
be dependent or independent of descriptive meaning, and it
may have the force of an invitation, a supplication, a com-
mand, or an impersonal rule.
But this is perhaps not so important as the next point to be
made. The fact of the matter is that the ceremonial theory
provides a special pattern of analysis which works very well
in certain situations but not in others. And, in the long run,
even in those situations to which it does apply it cannot be
fully understood if we fail to bear constantly in mind the non-
ceremonial context as well as the nonceremonial content of
9 Op. cit, p. 220.
the moral-judgment Let us consider the analogy be-

itself.
tween morals and which Miss MacDonald her-

religion to
self calls attention. Perhaps we will find that it is closer than
she realizes. Now as she describes moral discourse, it must
always be, in effect, "catholic." That is to say, it represents
a universal, public, impersonal authority which is wholly
independent of individual choice or preference. And the
acts which it enjoins or prescribes are distinguished as
"moral," not by the characteristic sentiments or feelings which
they may incidently express, but by the sacred forms or pat-
terns of the utterances themselves. In a word, it is not my
emotion which sanctifies the rite; it is the rite which sancti-
fies my emotion. From am
the standpoint of Catholicism, I
known as a Christian, not my motives or the
by the purity of
intensity of my private conviction but rather by my acquies-
cence in the public ceremonials of the Church. Just so, my
judgment is not rendered "moral" by the special "moral senti-
ment" which may animate it; rather, my moral motive is
known to be such because I express it in ethical terms. The
feeling does not, in short, invest the judgment with moral
authority or persuasiveness, but the judgment itself gives
the stamp of moral authority to the act of personal ap-
proval or disapproval.
But, just as the independent authority of a religious ritual
is gradually weakened when it ceases to be the effective
symbolic surrogate of collective social interests which "back
it up," so, in the ethical sphere, the moral judgment will tend
to loseits independent authority when it ceases to be re-
garded as the symbolic equivalent of the common demands

of a unified social order. The moving appeal of religious
myth begins to dry up when it ceases to express an inde-
pendent collective attitude or belief of the group as a whole.
And similarly, when the actual prescriptions that are ex-
pressed through the forms of the moral judgment no longer
express genuine social demands, the forms of the judgments
themselves will be incapable of long retaining their moving
appeal.
What is religious protestantism? It is, above all, the prac-
tical denial of the efficacy of general social and institu-

tional procedures in the matter of individual salvation.
It is inprinciple an assertion of the right of the individual
to employ the forms of religious utterance as an expression
of personal faith and private feeling. The religious author-
ity, so the protestant claims, does not lie in the ritual, but
derives exclusively from the personal conviction and piety
which alone give religious significance to the forms. What
protestantism means, therefore, from the standpoint of the
vehicular symbols by means of which the Christian, whether
catholic or protestant, affirms his loyalty to his religion, is
precisely a denial of the objective, public authoritativeness

of the symbols as such. The protestant continues to use the
symbols, but only because he can pour into them an expres-
sion of his own passion and devotion. This alone is what
gives them their authority for him.
Now just as the independent social authority of religious
ceremonials is inevitably weakened by the rise of protestant
religious attitudes, so, in the sphere of morals, it may be
argued that the authority of the forms of the moral judgment

lose their independent normative appeal when they are re-
garded as deriving their proper authority from the indi-
vidual conscience itself. And just as in the sphere of religion,
the emergence of protestantism marks a gradual breakdown
in the efficacy of the established ceremonials and their ac-
credited ministers, so the emergence of a spirit of protes-
tantism in morality results gradually in the breakdown of the
efficacy of the moral judgment itself as authoritative public
rite.
To generalize, I suggest that in the sphere of morality, the

rise of individualism has carried with it a gradual decline
moral judgment, and
in the special ceremonial status of the
a gradual weakening of the sheer persuasiveness of moral
expressions as such. Personally, I am inclined to think that
this may be a good thing, but I am here concerned to de-
scribe a fact, not to evaluate a tendency. If I am correct, then
just to the extent that an individualistic normative ethic be-
comes dominant, the interpersonal ceremonial theory of Miss
MacDonald and patterns of analysis similar

ceases to apply,
to Stevenson's working models come into their own. The ap-
plicability of the ceremonial theory presupposes, in short, a
people bound together by profound "organic" ritual and
ceremonial ties which are reinforced continually by inde-
pendent common concerns. But just to the extent that these
tiesare loosened, the moral judgment is transformed from
an impersonally authoritative ritual into an expression of
personal attitudes or decisions.
Another parallel between religion and morality may also
be instructive. It should be pointed out in the first place
that protestantism is not a wholesale repudiation of tradi-
tional religious formulae. On the contrary, although their
meanings are profoundly altered, the ritualistic patterns at
first remain precisely what they were. Thus, for example,
the protestant, like the catholic, continues to recite the

Lord's Prayer and to observe the ritual of baptism. The old
forms remain; they are simply invested with a new and more
intimate personal significance which emphasizes the feel-
ings of the individual rather than the common authority of
the group. "Our Father" becomes, in effect, "My Father,"
and the confessional is replaced by private meditation and
prayer. The verbal expression remains the same, but the
way in which it is said, and its effect upon the sayer is ut-
terly different.
Similarly in ethics, the repudiation of the ostensible author-
ity of the old social mores does not necessarily or even
usually involve a wholesale refusal to invoke the verbal
forms by means of which they were expressed. On the con-
trary, precisely because of their inherent normative appeal,
—
they are usually retained with a shift in their areas of ap-
plication.
But if the objects to which different individuals apply
the ritualistic terms of moral discourse are widely different or
opposed, so that in the end each moral judgment becomes
merely the private celebration of a personal moral rite that
is dishonored by one's fellows, then in the end the suasion of
the terms themselves will tend to disappear. The authority of

the moral judgment helps to solidify, but it also in turn de-

pends upon a sense of community with others. When it
ceases to be thought of as a symbol of a larger common in-
terest with which, in part, oneself is in sympathy, it loses
its distinctive moral status as something which can oppose
the inclinations of the individual as an impersonal or public

"ought." Hence, when the objects to which we attach or seek
to attach moral significance are widely different from those to
which the moral judgment is traditionally applied, it becomes
necessary, if the forms of moral discourse are to retain their
public authoritativeness, that we successfully redefine their
scope or areas of application so that henceforth they will be
applied both by ourselves and others to classes of objects of
which we already jointly approve on other grounds. And it is
at this point that the functional significance of Stevenson's

theory of persuasive definition in ethics becomes ethically
relevant.
However, it is of the first importance to observe that the
process of "irrigation" works both ways, and that the per-
suasive redefinition is quite as important as a means of "ir-
rigating" the ceremonial forms of moral judgment them-
selves as it is as a means of redirecting attitudes, through
the use of emotive language, toward new classes of objects.
As a rule no one attempts to redefine the scope of an
ethical term unless he also believes that the objects which
fall within that scope are more useful or agreeable than
those to which the term was previously applied. But unless

the objects to which the definition directs attention are
themselves already in the process of becoming socially de-
sirable, the definition will simply fail to be accepted. What
happens in the case of persuasive definition is, then, a two-
way modification of meaning, and a two-way modification or
reinforcement of attitudes. The ceremonial meaning of the
ethical term invests the objects to which it is reapplied
with the aura of special importance which the notion of
moral obligation tends to convey. But they, in turn, because
of our independent interest in them, reactivate the cere-
mony of the moral judgment.
ia6 Reason and Conduct
In general, I believe, too much attention has been paid

by analysts to the predicates of the moral judgment, and too
little to the subjects to which, in different ages, they have
applied. Not only do ethical terms have a meaning of their
own, but also they have conditions of sentential intersignifi-
* cance which limit the range of their application. These con-
ditions, I submit, are a function above all of common in-
i terests which, quite apart from their moral significance, are
of basic social importance.
To be sure, the range of application of ethical terms is al-
ways vague and variable. But indefinite as their range may
be, it is still true that there is at any given time a fairly
clear area in which we have no doubt as to the applica-
bility of moral terms, and another in which we have lit-
/tle doubts as to their inapplicability. These areas conform,
/ roughly, to those areas which are and are not, respectively,
matters in which the group as a whole is seriously interested.
When, as in our own time, these areas become confused,
either through indiscriminate redefinition or through their
persistent application to objects in which there is no inde-
pendent common interest, ethical terms will tend to lose
their authority.
Broadly speaking, then, the ceremonial use of ethical
terms cannot long remain effective where there are no un-
derlying common focal aims to which most members of
society are committed. Where the objects to which they are
applied are objects of indifference or revulsion, then not
merely will the moral ceremony lose its appeal but it may,
in the end, even become count er-incitive. Such a process is
sufficiently familiar in the sphere of religion, but it some-
times occurs also in the sphere of morality.
There are other considerations which alsotend to support
the present thesis. One of these is the well-known fact that
unless a conditioned response is at least occasionally rein-
forced by a correlated reduction of the underlying drive,
the stimulus gradually looses its efficacy to incite the re-
sponse. If this is so, then it would appear to follow that
the emotive appeal of any symbol will tend to fade un-
less the objects toward which it directs attention are them-

selves genuine satisfiers. Secondly, it appears that negative
reinforcement is generally far less effective as a means of
determining attitudes than is positive reinforcement. But if
this is so, then moral judgments which are reinforced merely

by a vague threat of social disapproval will tend to be less
effective as determinants of conduct than those which direct
the energies of the individual toward goals which are ob-
jects of some positive satisfaction.
This suggests that when the force of "morality" declines,
it may be due not so much to a general lack of seriousness
in the conduct of life, or to the perversity of certain moral
philosophers, but, rather, to the emptiness of a system or\
ceremonial rules" which have lost tneir positive social utility./
Whether men may live without "morals" I do not know.
I believe, however, that moral judgments cannot long re-
when the ends which
tain their public authority as incitors
they enjoin do not serve the permanent communal interests
of men.
[VII]
Global Conventions in Ethics
No one seriously doubts any longer that the full normative

sense of a moral judgment cannot be wholly explicated in
terms of its descriptive meaning, and that although such a
judgment may "contain" a proposition which is true or false,
its distinctive normative claim does not rest exclusively in
this fact alone. What was less evident before the clamor
over the emotive theory subsided is that there are conditions
of relevance in moral deliberation and argumentation. No
reason is a "good" reason just because it proves effective in
(sustaining attitudes prescribed by the moral judgment it-
self. There are, in short, governing normative conventions

in ethics which control our vagrant attitudes and correct the
associations that psychologically determine them. These con-
ventions are embodied in the notions of "normality," "reason-
ableness," "disinterestedness," "sensible person," and the like.
It is with the special characteristics and functions of these
conventions that this brief essay is concerned.
It is not profitable, I think, to debate the question whether
these expressions have any "cognitive" meaning. It would
be absurd to say that I do not "understand" anything by
them. But what it is that I understand is somewhat vague.
As we shall see, this does not in the least impair the useful-
ness of such expressions.
1 1 owe the term "global" here to Philip Blair Rice. See his On the
Knowledge of Good and Evil (New York: Random House, 1955),

which is discussed at some length in essay XII of this volume, pp.
240-62.
Global Conventions in Ethics 129
As in so many other places in moral philosophy, the ini-
tial insights are Hume's. Now most of Hume's critics have

classified him as a characteristic representative of the senti-^
mental school, according to which natural benevolence is the
source of all moral approval, and sympathy is the psycholog-
ical mechanism whereby benevolence is "irrigated." What
they have usually failed to observe is that an analysis of
moral judgments which reduces them to expressions of sym-
pathetic benevolence cannot account for the characteristic
"corrective" role which Hume assigned to "morals." Accord-
ing to him, we have no instinctual love of mankind as
such. Our natural sympathies are strongest in relation to
those who most closely resemble ourselves, weakest toward
those who appear most dissimilar. It is evident also, Hume
contends, that our sentiments vary in accordance with the
distance of their objects in relation to ourselves. Hence
our ordinary benevolent impulses and our individual acts of
praise and blame tend to be fluctuating and parochial. This,
however, does not greatly affect our moral judgments of hu-
man motives and actions.
In judging the moral character of any person or act, we
ignore our variable personal feelings and take a "more
general view." Yet we do this, not because of any particu-
lar decision on our part, but rather because of our habitual
use of certain impersonal social conventions which dispose
us to discount our variable temporary attitudes and "still
apply the terms expressive of our liking or dislike in the
same manner as if we remained in one point of view. Expe-
rience soon teaches us this method of correcting our senti-
ments, or at least of correcting our language, where the sen-
timents are more stubborn and unalterable." 2 "Such correc-
tions," he adds by way of comparison and elucidation, "are
common with regard to all the senses; and, indeed, it were
impossible we could ever make use of language or com-
municate our sentiments to one another, did we not correct
the momentary appearances of things and overlook our
2 Hume's Moral and Political Philosophy, edited by H. D. Aiken
(New York, Hafner, 1948), pp. 137-8.
3
present situation." It is also worth remarking in this con-
nection that Hume also speaks analogously of the corrective
role of aesthetic judgments which serve to offset our variable
individual responses to works of art. In the essay "On the
Standard of Taste" he points out how, despite the inescapable
truth of the dictum that there is no disputing about tastes, it
is also from a common-sense viewpoint absurd to deny that
there an order of merit among works of art which is some-

is
how public and independent of the variations of our

tastes. By this latter "species of common sense," he says sug-
gestively, the former is "modified" and "restrained."
These suggestions are immensely fruitful. At this point,
however, Hume leaves us to our own devices. Let us
compare, then, for what it may be worth, the concept
of "objective red" and analogous conceptions in ethics.
In both cases, first of all, a distinction is made between ap-
pearance and reality. In both cases a means is provided for
offsetting fluctuations due to the "accidental" factors af-
fecting our changing views of things. In both cases we are
enabled by the distinctions to make communications in
which the subject under discussion is responded to in similar
ways by both parties. But there are also important differ-
ences. The function of the judgment "x is really or ob-
jectively red" is to describe the appearance of x under cer-
tain standard conditions and to correct hasty beliefs based
upon appearances which may be abnormal. The result of such
a judgment, for the addressee, is information. He is thereby
enabled to predict how the object will appear to him if he
looks at it in a certain way. However, the function of the
moral judgment is not primarily informative and pre-
dictive. What it aims to correct is not our beliefs about mat-
ters of fact but our attitudes towards actions, motives, and
persons. It "restrains" our feelings, not our perceptions or, —
better, it determines us, in those deliberations upon which
our "general decisions" are based, to consider the object
imaginatively without regard to our more personal in-
terest. When we ask whether something is really or ob-
s Ibid.
jectively right, we are, in effect, enjoined to perform one of

the "imaginative experiments" so frequently employed in
books on moral philosophy. More than this, a type of at-
titudeis appealed to which regards the objects reflected
upon from a certain point of view, which we call "disin-

terested" or "social." We are induced, so to speak, to im-
personate the role of a person imbued with only the funda-
mental social presses of the community. The effect of ethical
discourse is thus to standardize and coordinate our delibera-
tions and general decisions, just as the effect of statements
about objective predicates is to standardize and coordinate
our beliefs about matters of fact.
In ethics, several normative conventions prevail which, al-
though they do not precisely correspond to any actual indi-

vidual sentiment, nevertheless serve to induce certain rela-
tively uniform attitudes in most persons to whom they are
addressed. They serve to "correct" our sentiments, and yet
"they" are also dimensions of our natures; otherwise their
normative appeal would be nil. It is, after all, only because
we are social animals that they retain any normative hold
upon us.
There is another difference between "really red" and
"really right." Both, as we have seen, involve notions of "ob-
jectivity," "competence," or "reasonableness." But in the
latter case, the conditions of normality cannot be precisely
stated.Perhaps the difference is one of degree; if so, the de-
gree of vagueness in the latter concept is far greater. But if
we bear in mind the different roles of the respective concepts,
we shall see that the vagueness of "competence" in ethical
criticism is proper to it. The point is that the functions of ethi-
cal criticism are (a) to dispose us to reflect deliberately and
dispassionately, not to predict precisely what will occur, and
(
b ) to produce agreement in our general decisions and in our
procedures for resolving lower-order disagreements that in-
evitably arise in applications and criticisms of the prevailing
moral code. For such purposes the operative concepts need to
be flexible and vague. Too much rigidity in meaning would
render them practically useless. The "realities" with which
ethical criticism is mainly concerned are "sober second

thoughts" and "calm passions," the settled convictions which
determine our long-range behavior as moral agents. It has
no particular objective or goal in view, merely a way of
settling differences that arise from our particular passions
and preoccupations.
In speaking of "global conventions" in ethics I do not mean
"decisions" as this term is ordinarily used by philosophers. I

may decide to go to the grocery store; I may decide to let
the word "value," in a particular essay, mean "object of de-

sire." But just as the ordinary meanings of words are not es-
tablished by decisions or stipulations, but rather by uncon*
scious processes of association and imitation, so the conven-
tional principles that govern and correct our moral reflections
and deliberations are learned ideo-motor dispositions, not
particular "acts of will." This is part of what is meant in say-
ing that ethical norms are "social": they are acquired by
processes of social conditioning and imitation of which we
are largely unaware.
It may be worth while, at this point, to compare the role
of global ethical conventions with what recent writers have
misleadingly called "myths" in the sphere of political ideol-
ogy. Both expressions may confuse the unwary. Just as the
former term may mislead the unwary into supposing that
ethical norms are personal decisions, so the latter term may
suggest to some that a political myth is a deliberately per-
petrated he by means of which the rulers control the ruled.
But Harold D. Lasswell and Abraham Kaplan
as Professors
observe in connection with their own discerning analysis of
political myths, the expression need not be interpreted as
necessarily imputing a fictional, false, or irrational character
to ideological symbols. 4 As they put it, "myth" refers to an
aspect of "their functioning" not "their properties." 5
A politi-
cal myth does not describe any existing or historical state
of affairs. Nor does it prescribe any particular political act or
4 H. D. Lasswell and Abraham Kaplan, Power and Society ( New

Haven: Yale University Press, 1950), pp. 116 ff.
5 Ibid., 117.
p.
obligation. Rather, its role is to set in motion and to direct
certain general attitudes toward existing political institutions.

But whereas myths function, as Lasswell and Kaplan point
out, in preserving or supplanting the existing political order,
ethical conventions, as here conceived, have a dual role.
Since they permit the "objective" criticism of existing moral

rules, they do not serve to preserve the existing moral order
altogether. But since they themselves involve impersonal
standards of propriety or reasonableness, they are not cal-
culated to produce a complete or revolutionary transforma-
tion of lower-order moral attitudes already in operation. In
short, they neither wholly supplant the existing moral code
nor wholly sustain it.Rather do they enable its equitable cor-
rection or modification within a broader framework of gen-
eral norms which itself changes slowly by processes of ac-
cretion and reinterpretation.
I suggest, then, by way of conclusion, that the role of such
notions as "reasonable" in ethics is to dispose us (a) to take

a "second look," to consider a wider variety of alternatives
and a longer range of consequences than we have hitherto
considered, and (b) to judge without regard to our present
or m erely personal feeH ngsTTfie^h|mictioii "Be reasonable"
or "Be_objective," in ethics, does not tell us what to look
for or what to approve. It does not prescribe any particular
attitude of favor or disfavor. Instead, it induces a disinter-
ested point of view or press which invites imaginative ex-
periments and the impersonation of the role of the "social
man." For no two persons will the point of view have pre-
cisely the "same" meaning. For no person will its meaning
be exact. But again, this is not a serious handicap, given
the function which such concepts perform.
[VIII]
The Concept
of Moral Objectivity
INTRODUCTION
What are ordinary persons, including philosophers in their
ordinary moments, doing when they raise doubts about the
objectivity of particular moral judgments? And how, as
moral agents and critics, do they go about resolving such
doubts? These questions, one would think, must have oc-
curred to anyone who in an idle moment bethought him-
self about the nature of moral objectivity. Yet they seem
rarely, if ever, to be considered by the ethical objectivists and
their critics, in spite of the interminable debate between
them —or more likely, because of it. In the case of the sub-
jectivists this is hardly surprising, since it would be fatal to
their cause to admit that such questions may be seriously
raised at all. But the objectivists make no use of this advan-
tage; in fact, they manage merely to create the impression that
the notion of moral objectivity is question-begging. The rea-
son for this is evident: no objectivist has examined the use of
the concept of objectivity in moral contexts without precon-
ceptions about its generic meaning derived from a continu-
ing philosophical tradition for which formal logic and natural
science have provided not only exemplary, but paradigm
cases of objective discourse. Suppose that we challenge these
presuppositions. Why, be assumed in
for example, should it
advance of analysis that moral judgments cannot be objec-

The Concept of Moral Objectivity 135
tive unless such words as "right" and "good" are terms of

"objective reference," or unless such judgments are true or u^
false statements about something called "objective reality"?
For that matter, why should we not take that reality to be
whatever, in context, answers to our objective questions?
Wfa^_should we assume that proper application of the c on-
cept of ob jectivity in all contexts involves the notion of a
consens us of "competent," "rational " nr "idpal" o b< p rvpr<;? ; .
And, if we do not, why must we assume that meaningful ap-

plication of the concept involves definitive disciplinary pro-
cedures or rules to which anyone who sets up for a moralist
is logically bound to submit?
When I first came within range of the ideas underlying
the preceding remarks, they at once appeared to me to open
a path through a swamp of controversy to a firm and neutral
ground where I could proceed without ado to a constructive
analysis of the concept of moral objectivity. The questions
raised at the beginning of this essay seemed completely
irenic; they were calculated, so I thought, not to increase
pre-existing philosophical doubts about the possibility of ob-
jective moral judgments, but rather to allay them. For if
these questions can be asked at all, then such doubts, as
well as the controversies to which they give rise, must surely
be both gratuitous and perverse. The implication, plainly,
is that instead of endlessly arguing whether, in principle,
objective moral judgments are possible, philosophers would

do better to ask themselves how a serious problem about
the objectivity of morals could even arise. But there, pre-
cisely, is the rub. Whether or not my questions are neutral,
they at any rate are not pointless. Why not? Let us consider
an analogy. Why is it that since Plato there has been no
philosophical interest or stake in the concept of mud? To be
sure, lexicographers have asked (and to their satisfaction,
have quickly found answers to) the question "What is the
meaning of 'mud'?" But among philosophers, only Plato has
ever thought of asking that question, and he himself did so
merely in order to illustrate a point, without staying for an
answer. The philosophical interest in the concept of moral
objectivity is am not trying

of another sort. In this essay I
to illustrate a point at but rather to resolve genuine
all,
philosophical doubts, not only about the meaning, but also

about the meaningfulness of the concept of moral objectiv-
ity. The fact is that philosophers, unlike lexicographers, never
inquire into the meaning of any concept in a purely specula-

tive frame of mind. For them there are always two ques-
tions to be asked: (a) "What is the meaning of V?" and
(b) "Why ask such a question?" This means that within a
philosophical context, no question about the meaning of a
word, or the use of an expression, is ever purely irenic,
and no answer to it is significant until it is shown how it
bears upon the perplexities, at once intellectual and practical,
that dispose us to ask it. Indeed, it is only its power to re-
solve such perplexities that convinces us that the answer
itself is substantially correct. The trouble is that the per-
plexities themselves prevent us from seeing where the an-
swer lies.
These remarks apply directly to the case at hand. There
would be no philosophical reason to undertake an analysis
of the concept of moral objectivity were it not for the wide-
spread, morally destructive doubts about the meaningfulness
of such a concept. Such doubts may once have been gra-
tuitous; they are no longer. On the contrary, philosophical
preconceptions about the nature and conditions of any ob-
jective discourse have, in one form or another, so condi-
tioned our thinking that laymen, as well as philosophers,
take it for granted that rejection of ethical objectivism auto-
matically commits us to the thesis that, despite appearances,
moral objectivity is an illusion and hence that the ordinary
language of morals is systematically misleading. Thus, those
who accept the responsibilities of objective judgment in
morals have been impelled to reinstate their own principles
as laws of nature or as definitive principles of morals which
cannot be challenged without declaring either one's immoral-
ity or one's incompetence as a moral being. On the other
hand, those who regard all such positions as morally un-
tenable, on the ground that they profoundly jeopardize the
—
principle of moral autonomy or freedom, seem thereby com-

mitted to some form of moral subjectivism which precludes
one, on principle, from asking whether one's judgments and
principles are objective. Meanwhile, our ordinary moral prac-
tices are hobbled for want of a clear and morally suitable
conception of their critique, and the whole institution of
morality gradually acquires the appearance of a system of
arbitrary dicta which have not yet attained the status of
positive laws. The moralist is demoted to the rank of a
busybody who has fortunately not yet discovered his voca-
tion for politics, and the immoralist assumes the status of a
cultural hero whose bad conscience is treated as a badge
of courage. The rest of us —which, I suspect, includes prac-
tically everyone for whom neither immorality nor morality
is a vocation —find ourselves, when we reflect, involved in
a practical dilemma which forces us to vacillate perpetually
between the bad conscience of objectivism and the equally
bad conscience of subjectivism. Our unhappy situation is
this: we wish to honor, indeed, we cannot escape, the obliga-

tion to be objective in our moral decisions, yet we seem
unable to do so without at the same time committing our-
selves to a conception of morality which, if taken seriously,
destroys our autonomy as moral agents and critics. On the
other hand, we find ourselves committed to a principle of
moral freedom which apparently dooms us to acquiesce in a
radical ethical subjectivism that renders meaningless the
very effort to search for objective moral judgments. This
predicament, let me emphasize, is not merely theoretical
it is practical; or better, the problems with which it con-
fronts us are not speculative problems about moral discourse,

but theoretical problems that arise within it whenever we
are obliged by moral necessity to go to the fundamental
principles underlying our moral practices. Until they are
— —
solved or resolved judicious men who wish to view
moral reflection as a part of the life of reason must con-
tinue to have grave, even paralyzing, doubts about the
mutual consistency of the basic critical practices from
which our familiar notions of the moral life are ultimately
derived. Thus, although I still take seriously the questions

posed at the beginning of this essay and hope presently to
find answers to them, it would be philosophically pointless
to proceed to a constructive analysis of the concept of moral
objectivity without regard to the philosophical perplexities
which provide the only reason for understanding it. This
delay, however, is not without its reward. As I have found,
it is precisely by facing these perplexities that we come
gradually in sight of a conception of moral objectivity to
which, without illusion, practicing moralists as well as moral
philosophers can give credence.
My first task, then, will be to spell out the ethical antinomy
which gives rise to the question of how objective moral judg-
ments are possible. My second task will be to find a way out
of this antinomy which, without compromising what seems
essential to the principle of moral autonomy, nevertheless
preserves intact the basic minimal claims embodied in the
principle of moral objectivity. Such a solution, of course, is
possible only at a price. In this case, fortunately, the price is
not exorbitant; indeed,can be readily paid without in the

it
least compromising what remains of our common moral

sense. The conclusion I am forced to is that, at least so far as
ethics is concerned, traditional philosophical notions con-
cerning the meanings of the terms "objective" and "subjec-
tive" must be abandoned once for all, and that neither
empirical science nor formal logic can any longer serve
philosophers as the models of objective discourse. I shall argue
that it is precisely the uncritical application of such notions
and the noetic models from which they derived, beyond the
domains of science and logic, that is ultimately responsible
for the widespread moral skepticism with which at present
most of us are afflicted.
I shall attempt, in particular, to show how groundless is
the suppostion that there can be objective moral principles

only if some universal standard or principle of moral
there is
and wrong which is acknowledged as binding by all

right
men of good will. Commitment to the ideal of moral ob-
jectivity does not entail the skeptical conclusion that no prin-
ciple can be regarded as morally valid unless it can be

viewed as binding upon every "competent" moral agent; nor
does it commit us to the impossible thesis that the funda-
mental responsibilities and rights of all moral persons must
be substantially the same. Quite the contrary. There is no
such thing as a competent or incompetent moral agent; in-
deed, the supposition that there is, is itself a sign of moral im-
maturity. The ideal of moral objectivity must be adjusted to
the possibility of an essential diversity of moral codes and
not merely to an accidental diversity of moral opinions.
There can be no such thing as "the moral point of view," and
if any supposed rule, whether linguistic, legal, or theological,
were to serve as a moral principle, it is only we as individual

moral agents who could make it do so. If the maxims or
precepts that now serve many of us as moral principles
should ever come to be understood as definitive principles
of morals, then the whole notion of moral agency as it is
now understood would simply disappear. But it is only when
the implications of this fact have been fully appreciated that
we can adjust ourselves to the ordinary notion of moral ob-
jectivity whose meaning it is my eventual purpose to ex-
plain.
THE ANTIMONY OF MORAL

OBJECTIVITY AND FREEDOM
The Presumptive Principles of Moral Objectivity. In prin-
ciple, it is always proper to inquire of any moral judgment
whether it can be objectively sustained. It is proper, also, to
ask of most moral principles whether they are objectively
valid. But no such judgment or principle can be regarded
as objectively valid unless there are certain definitive prin-
ciples of morals which are binding upon the judgment and
conduct of every moral agent. It does not suffice, as Kant at
times seems to imply, that every moral agent must be ready
to treat the maxim of his judgment as the principle of a uni-
versal legislation, for this plainly leaves open the possibility
that conscientious men could consistently differ on all mat-
ters ofmoral principle. Objective moral principles are possi-

ble only if there is at least one universal principle of morals to
which every moral judge and agent is beholden in justify-

ing particular moral judgments and lower-order principles.
Since one of its essential functions is to guide our conduct in
our interpersonal dealings with one another, morality cannot
be regarded as a positive science; so much may be conceded
with impunity to those so-called "non-cognitivists" who mis-
takenly deny that moral judgments are true or false state-
ments. However, this does not imply that in ethics anything
goes, or that there is no such thing as a discipline of morals,
by appeal to whose rules it can in principle always be de-
cided whether any moral judgment is true or false. Like
positive science, morality must be understood as a universal
discipline to which every moralist must submit. And it is
for this reason that we may speak, in ethics as in science, of

"qualified observers."
For the sake of clarity, it is well to point out certain conse-
quences of these principles which moral objectivists usually
fail to acknowledge. Morality, as we have seen, must be re-
garded as a discipline. However, every featherless biped

must decide for himself whether he should submit to such a
discipline. In this respect, morality may be likened to a game.
That is to say, any man must act in accordance with the
rules of the moral "game" if he is to "do" morals at all. But
there can be no moral rule which makes it necessary for
every human being to play the game of morals, or if there
were, then it would be up to every individual to decide for
himself whether he wishes to be regarded as a "human be-
ing." Most of us are forced by the circumstances of social
life to submit to the discipline of morals, or at least to give
the appearance of doing so. But other ways of handling one's
personal relations are clearly conceivable. And if an in-
dividual is willing to risk being unpopular with the moral-
ists, he is free not to play, or, having played for a term, to
play no longer.
There are other consequences of the principles of moral
which many will find more agreeable. For one
objectivity
thing, if there are certain universal and necessary principles

of morals, then in principle every contingent moral disagree-
ment can be rationally resolved. Accordingly, every such
disagreement may be properly understood as merely a dis-
agreement in opinion rather than a disagreement in prin-
ciple. Owing to the intrusion of such subjective factors as
self-interest, stupidity, and ignorance, such disagreements
may be in practice extremely difficult to settle. Nevertheless,
we must assume that there is at least a tendency toward
agreement on the part of conscientious moralists as the facts
come more fully into view and as they become more fully
aware of the nature of their commitments as moral beings. In
short, underlying every substantive moral disensus there is a
basic consensus which ensures the possibility of a rational
resolution of any moral disagreement that may arise.
It should be added that when moralists pass beyond a cer-
tain point in their disagreements, the issue between them
becomes a concealed verbal dispute over the meaning of the
concept of morals itself. This fact helps to explain why, in
certain instances, the controversialists tend to go round and
round without being able to compose their differences. When
such situations arise an objective solution may still be
reached, but only by appealing to the rules governing the
common use and application of the concept of morals it-
self. Should doubts persist past this point, then the reply
must be that they involve a skepticism about the language

of conduct generally which can hardly be sustained apart
from a skepticism about the rules of ordinary language as a
whole. But in that case, the issue has already passed outside
the domain of moral philosophy into a sphere where the
very notion of an objective disagreement is largely mean-
ingless.
The Presumptive Principles of Moral Autonomy or Free-
dom. According to the preceding conception of morals, the
principles of morals, like the rules of a game, are only hypo-
thetically binding upon the individual person: if anyone
elects to do morals, then, as a moralist, he must conform to
its disciplinary principles. In such a view, accordingly, the
individual acquits himself morally so long as his actions ac-

cord with such principles, regardless of his personal reasons
for conforming to them. In other words, as a moral being
anyone is ultimately responsible only to the principles of
morals, not to his own conscience. To be sure, a man may be
excused from responsibility for wrong actions if he con-
scientiously performs them in the belief that they are right;
but such an excuse makes sense only if (a) the moral agent
has (mistakenly) judged that his action is, on the evidence,
the objectively right thing to do and ( b
) there is in the situa-
tion an objectively right thing to be done which others may

know as well as, or better than, he.
This view of morals is incompatible with the principles of
moral autonomy or freedom. According to these principles it

is not enough that the moral agent should be capable of mak-
lg mistakes, nor is it enough that he should be free to

violate the moral law. As a moral being he must, also, in
principle, decide absolutely for himself what that law really
is. As we sometimes loosely and misleadingly say, every
^genuine moral agent must be regarded as a "law unto him-

self." That is to say, no man is morally responsible for actions
unless they are performed for the sake of principles which he

cannot in conscience disavow. Here we do not just excuse a
man who acts on principles which, as we may think, are
objectively wrong. On the contrary, if he sticks to his prin-
ciples "though the heavens fall," he is entitled to our respect
and perhaps even to our admiration. It is for this reason that
morally sensitive men not only forgive those who conscien-
tiously oppose them on the ground that the latter "know not
what they do"; when the issue between them is a matter of
principle, they generously, if also tragically, honor them for
exhibiting a moral integrity as great as any to which they
themselves may aspire. A more sensitive Antigone would
understand that she cannot totally condemn Creon without
at the same time condemning herself.
The fundamental point is that morality cannot properly
be regarded either as a form of law, as a book of rules, or
as a set of socially authoritative commands. It does not
suffice, as the objectivists maintain, that a man may be ex-

cused or forgiven if he happens unintentionally to do what is
objectively wrong. For who is entitled to excuse him and who
is in a position to grant him forgiveness? In morals there is
and can be nothing to do save to follow the principles to

which, upon reflection, one finds oneself committed. Such
principles present them gP> lv p s in conscience rs categorical
imperatives, not as laws, rules, or commands which are
bin ding upon us if we are pleased to do the moral thing. But
just because of this it makes no sense, in the moral sphere, to
speak of a discipline of morals or of the principles of morals;
here, indeed, we can speak only of "my principles," "our
principles," "his principles," or "their principles." And the
ordinary language of morals must and does accommodate
itself to this fact.
In the domain of science and logic, where we can speak

with a straight face about disciplinary principles or rules,
situations often arise in which we properly defer to the
authority of observers whom we recognize to be more com-
petent or qualified than ourselves. And it is because of this
that, without qualms, we accept certain statements as ob-
jectively trueeven though we ourselves do not fully see why
they are But in morals such situations can hardly arise.
so.
For just as no one can live by another's principles, so no

one can be expected to conform his judgment and his will to
certain allegedly objective principles which he has not in
conscience made absolutely his own. Nor is this situation
altered by the fact that some men take their principles
from some "authority." For that authority can make no moral
claims upon anyone who does not adopt it as his authority.
In short, while a man may adopt as his moral principle,
"Always act in such a way as would meet with the approval
of (say) the church," that principle is no less a personal
precept than the principle of utility or the principle of verac-
ity.
The principle of moral autonomy can now be restated in

the following way: Every moral principle must be regarded
as nothing more than a first-personal precept. To be sure,
such precepts may be either singular or plural. But no matter

how extensive may be the community which commits itself
to a particularmoral practice or principle, it is morally bind-
ing only upon the members of that community. The prin-
ciple of moral autonomy is thus incompatible with the very
notion of a universal discipline of morals to which the con-
science of every moral person is objectively beholden. Hence,
so far at least as the concept of objectivity depends in
principle upon the idea of an underlying consensus of "com-
petent" or "qualified" moral judges, it has no application
within ethics. To that extent there can be no such thing as a
principle of moral objectivity. Moral discipline is merely a
personal regimen, or way of whose character is de-
life,
finable only in terms of those precepts to which the in-

dividual moral agent holds himself responsible.
By the same token there can be no paradigm cases of a
moral principle which every person who understands the
language of morals is bound to accept. For if there were
such a principle, then every moral agent would automatically
have to regard it as binding upon his own judgment and
conduct. But the principle of moral autonomy itself pre-
cludes such a possibility, even in the case of the principle of
compassion itself. In rejecting the principle of compassion,
Nietzsche, for example, did not thereby declare himself to
be going beyond moral good and evil but only to be "trans-
valuing" the principles of what he considered to be a "slave
morality." Those who really go beyond moral good and evil do
so for one reason only: because for them the whole idea of
—
moral obligation like the idea of God for some others is —
simply "dead."
It follows from the principle of moral autonomy that
meaningful disagreement is possible only among the mem-
bers of a particular moral community. So-called "cross
cultural" moral criticism can only be regarded as a kind of
propaganda, the main function of which is to reassure those
"at home" of their moral rectitude. At this level the distinc-
tion between "justifying" and "exciting" reasons has no appli-
cation. And those who talk, philosophically or theologically,
of an objective moral law, natural or otherwise, delude no

one but themselves.
In the light of these remarks we begin to see the use to
which the autonomist is likely to put Moore's "open ques-
tion" argument. In conscience we are morally free to question
any alleged principle of morals. In order to assert our auton-
omy as moral beings we may even be obliged, in certain
circumstances, to defy it. This, so far as I can see, is the
principal point concealed in Nietzsche's "transvaluation of
values." Its significance lies not so much in the fact that it
rejects the principle of compassion, but that it asserts the

right of autonomous moral agents to reject any principle
when it is presented authoritatively as the principle of
morals. In short, from the standpoint of the principle of
moral autonomy, ethical naturalism in any form and on
any level is morally subversive.
TOWARD A RESOLUTION
OF THE FOREGOING ANTINOMY
Let us now examine, somewhat dialectically, certain internal
weaknesses in each of the preceding views. For convenience
I shall henceforth refer to them as "objectivism" and "auton-
omism." As we proceed we will also consider reformulations
of them designed to remove such weaknesses. In so doing it
is my aim to show how we are driven at last to a conception
of moral objectivity radically different from that to which

objectivism itself appears to commit us. This conception, as
I believe, is entirely compatible with at least one form of
the doctrine of moral autonomy. But first we must overhaul
that doctrine itself.
Now a thoughtful objectivist may point out that autono-

mism, as it stands, is internally inconsistent and that in re-
moving this inconsistency the autonomist must acknowledge
the basic minimal claim of objectivism But when this
itself.
claim is admitted the case for autonomism weak- is radically

ened, since there are other principles whose claim to be re-
garded as definitive of morality is prior to that of moral
.
autonomy. This may be seen in the following way: The

autonomist asserts that there can be no definitive principles
of morals, since, if there were, no moral agent could be
regarded, as he must be, as a law unto himself. But if
every moral agent must be regarded as a law unto himself,
then the principles of moral autonomy must be themselves
viewed as disciplinary principles of morals. In that case the
autonomist must admit on principle that there is at least one
principle by appeal to which particular moral judgments and
principles may be objectively verified ( or falsified)
The autonomist may reply to this simply by taking the
bull by the horns. He may agree that his position, as pre-
viously stated, is inconsistent, and what he should have
claimed is that since the principles of moral autonomy are
definitive principles of morals there can, by the nature of the
case,be no other such principles. The principles of moral
autonomy are, so to say, principles to end all ideas of morality
which dispose us to treat our moral precepts as objective
laws or rules of conduct. These principles are not only a
guarantee against moral presumption, but an absolute char-
ter of —
moral freedom. Such a view let us call it "essen-
tialistic —
autonomism" obviously commits its proponent to a
definitive closing of the open question so far as the con-
cept of morals is concerned. But this (so it may be said) is
no cause for alarm since what it closes is only the question
concerning the nature of moral judgment or of moral right
and wrong. As such it does not in the least run afoul of
—
Moore's naturalistic fallacy argument which holds so far as
it does hold —
only against descriptive definitions of such
words as "good," "right," and "ought."
It may be questioned, however, even with respect to the
concept of morals, whether the open question can be closed
without violence, or if it can be, whether the autonomist him-
self is in a position to close it. For the sake of argument let
us assume that in principle it can be closed. But definitive
principles of morals should be able to meet the test of
counter-instances. This, it may be argued, the principles of
moral autonomy cannot do. It is very easy to imagine a seri-
—
ous person who would disallow any alleged obligation as a

moral obligation if its fulfillment involved the performance of
an unjust act, regardless of the fact that it was performed
in accordance with precepts conscientiously avowed by the
agent himself. Of course, a sufficiently resolute autonomist
might deny that such an example provides a true counter-
instance. His only mistake, in that case, would be to suppose
that he thereby secures his own position. For in thus privileg-
ing the principles of moral autonomy he thereby automati-
cally converts them into a statement of the conditions under
which alone he himself is prepared to judge an act as morally
obligatory. In one sense, such a statement may be regarded as
a definition. But what it defines is merely a particular style
of moral judgment. It leaves the concept of morals itself un-
touched.
Suppose, now, that the autonomist sees that his doctrine
cannot plausibly be defended in its essentialistic form. That
is he at last realizes that the principles of moral auton-
to say
omy can be regarded as formal principles only in the sense
that they define what he himself accepts as a standard of
"good form" in matters of moral judgment. He also sensibly
acknowledges that the only reason he can disallow his critic's
supposed counter-instance is that it really functions in effect
only as an exemplary counter-principle. In a word, the au-
tonomist now frankly regards the principles of moral auton-
omy themselves as first-personal precepts which serve to de-
fine only his own moral point of view. What they prescribe,
so far as he is concerned, is that the conscientious judgment
of every moral agent ought (not must) absolutely to be re-
spected and that no action performed for the sake of the
principles embodied in such judgments is morally censur-
able.
Such a position — let us call it "preceptive autonomism"
is perhaps not inconceivable, however hard it may be in
practice consistently to maintain it. There are, however, a
great many prima facie duties which ordinary men, includ-
ing most autonomists, do not question in their dealings
with their fellows. The principles of moral autonomy re-
quire the autonomist to respect such duties in others; within

the limits of such respect they permit him to acknowledge
such duties for himself. But they provide no standard for
judging which duties are to take precedence when they con-
flict with one another. And they provide but one principle
for making the legitimate exceptions which, in practice, vir-
tually every moral principle is subject to. In practice, I sup-
pose, most men would hold that a promise may properly be
broken if keeping it involved a great deal of suffering to
(innocent people. But the principle of moral autonomy itself
i break a promise only when the keeping of it

entitles us to
ynfringes the moral freedom of others. Many men would
probably agree also that it is proper to tell a lie in order to
give comfort to a person at the end of his tether. But the
principle of moral autonomy provides no basis whatever for
making such proper exceptions to the principle of veracity.
In short, while the principles of moral autonomy provide a
final restraint in judging the conduct of others, and a mini-
mal basis for making exceptions in the case of our own prima
facie duties, they otherwise leave individual conscience com-
pletely without guidance.
It is precisely at this point that the objectivist has another
inning.For the autonomist, like anyone else, has the prob-
lem of deciding what he really ought to do when a conflict
of duties occurs within the limits set by the principles of
moral autonomy. How is
such a conflict to be resolved? It is
entirely possible, moreover, that a particular duty, such as
compassion, may oblige him to press beyond those limits.
What then? Since the principle of moral autonomy is itself
merely a precept, on what basis is the autonomist to decide
to adhere to the principle of autonomy rather than to
follow the obligations of compassion? He may reply that he
needs no basis for deciding, since, for him, the question is
already settled. But then he must admit that his decision is
completely arbitrary and that, in good conscience, it might
just as well have gone the other way. Next week he may find
that it does go the other way; he will then have just as much
or just as little reason for the line he takes. But the situation
is even worse than Suppose that the autonomist sticks to

this.
his principles and without exception. For him

absolutely
there is no moral conflict between the principles of moral
autonomy and other principles since the latter simply give
way whenever the limits imposed by the former have been
reached. But what are those limits, and how are they to be
found? It should be borne in mind, as Aristotle long ago re-
minded us, that ethical concepts have not the same exacti-
tude as those employed in logic or science. Here judgment
is required if they are sensibly to be applied. But what shall
guide the autonomist's judgment when he has to decide

whether a prima facie duty has trespassed upon the moral
autonomy of other persons? Indeed, it may be argued that
the very admission of the necessity of judgment in applying
the principles of moral autonomy involves another princi-
ple more absolute than autonomy itself. But even if the
autonomist refuses this gambit, the problem of boundaries re-
mains a source, not just of theoretical perplexity, but of con-
stant practical moral doubt. For the frontiers of moral
autonomy are nowhere clearly marked, and the petitions of
supposedly lower-order principles for satisfaction of their
claims arebound to be an everyday occurrence. Beyond such
commonplace harassments there also remain more general
problems of application to which the principles of moral
autonomy themselves provide no clue. For example, does the
principle ofautonomy entail respect for those individuals
who profess no moral principles and whose lives are dedi-
cated exclusively to their "work" or to their own pleasures?
Does it entail respect for those whose only conscientious aim
is to subvert the whole enterprise of morality? Do the prin-
ciples of moral autonomy permit us to attempt the re-
education of those who are not themselves autonomists, or
does the respect we owe to moral persons require that we ac-
cept every such person once and for all just as we find him?
Above all, who shall count as a "moral person," and when
does he attain his majority? These problems are not external

problems pressed upon the autonomist from a different
moral point of view. They are wholly internal to autonomism
and pending the answers that are to be made to them,

itself,
autonomism remains, morally, a merely abstract entity.

From the depths of the autonomist's own conscience there
comes the demand for some way of knowing what autono-
mism really is, what it really commits him to.
In another domain this is precisely the predicament of
Descartes, whose invocation of the principle of intellectual
autonomy forced him at the same time to look within his
own consciousness for a way to distinguish subjective ap-
pearances from objective reality. "But surely," the objectiv-
ist will reply, "this is just what cannot be done. Nothing will
come of nothing, and if Descartes claims absolute freedom
to think for himself, then whatever rules of method he
finally adopts can be nothing but subjective precepts. The
reality' which he claims to discover by their application re-
mains a merely subjective reality." It is here that the objec-
tivist may hope to make his stand. But let us now have a look
at the internal problems in which his own position is in-

volved.
In the first place, as the autonomist will quickly point out,
the very which we observed in the case of essen-
difficulties
tialisticautonomism also beset every objectivist at every
turn. For no matter what principles of morals the objectivist
comes up with, he must submit his theory to the test of
counter-instances. If, when a counter-instance is seriously
proposed he refuses to accept it as a counter-instance, then
he shows by that very fact that his principles are definitive,
not of morality as such, but only of his own style of moral
judgment. But if he accepts it, his theory is overturned. Is
there any such theory that remains beyond the reach of
counter-instances? Or, how are we finally to determine just
what the principles of morals really are? The question is
exigent since, unless actual principles of morals can be

supplied, objectivism is merely a vacuous possibility. In a
word, objectivism can be shown to be possible only by con-

fronting us with the unassailable fact of definitive principles
of morals to which no one who understands what is being
said can seriously make objection. This is a tall order.
There are only two alternatives. One way is for the objec-
tivist himself simply to take the bull by the horns and to tell
us, without ado, what really are the principles of morals.

But this venture will not do, unless every moralist, without
exception, acknowledges that every action prescribed by
the principles in question must be morally right. None of the
traditional forms of objectivism has managed to meet so
stringent a test. We have already observed the difficulties in
which the objective autonomist finds himself. Counter-
instances —
if that is what they are —
have been proposed to
every known form of theological objectivism. The only theory
known to me which seems to stand a chance of survival is
that form of humanitarian ethics which claims that what we
mean by a morally right act is one which by intention is com-
passionate, i.e., which seeks, so far as possible, to relieve
suffering. The difficulties involved in such a view are notori-
ous. For example, there are many conscientious moralists
who would accept acts of compassion as morally right only
if their performance does not violate the personal integrity of
the individuals involved. At best, the compassionist may
argue that no act is morally right which involves unnecessary
or needless suffering. But, what is to count as "needless
suffering," and who shall decide? There are also moral obliga-
tions which, as such, have nothing to do with questions of
suffering at all: for example, the obligation to be fair in one's
dealings with others, the obligation to tell the truth, or the
obligation to keep promises sincerely made. Here the prin-
ciple of compassion itself works primarily as a principle of
exception which, by that very fact, presupposes that there
are, in principle, forms of moral obligation which are non-
compassionate.
It may be argued, however, that we have hitherto failed
to distinguish sharplyenough between material or substan-
tive moral principles and formal or definitive principles of
morals. But where is the razor's edge which divides them?
The principle of compassion, let us agree, is a substantive
principle like the principles of autonomy. But then, if we
are objectively to resolve disagreements about the commit-
merits which these principles must be certain

enjoin, there
other principles which really do what is meant by a
define
"morally right action," and which, because of this, state what
every moral agent or judge must consider in trying to decide
whether adherence to such substantive principles is morally
right. What are these purely formal principles?
It is at this point that the other alternative open to the ob-
jectivist appears most appealing. He may now suggest that
our have been due precisely to the failure of the
difficulties
older objectivists to be sufficiently abstract. Suppose, how-
/"ever, that the very concept of ob jectivity itself is vi ewed ja.s
f a formal standard of moral right and wrong. In snort, what
\ if it were argued simply that all and only those acts are
(morally right which would be approved by anyone who
}views them objectively? Whatsis, wanted is a definitive state-
(ment of the characteristics of an objective observer.
MORAL OBJECTIVITY AND THE IDEAL OBSERVER

Now it turns out that this alternative has already been care-
fully explored by Professor Roderick Firth in his essay,
x
"Ethical Absolutism and the Ideal Observer." Accordingly,
by examining it in some detail, we may, I think, take the
final measure of objectivism in its most rigoristic form.
As Firth is aware, his analysis is not entirely without
precedent. Read in one way, it may indeed be viewed as an
attempt to clarify the doctrine of a disinterested or impartial
spectator which, in one form or another, is to be found in
the writings of Hutcheson, Hume, Adam Smith, and Kant.
Firth takes these writers to be arguing, in effect, that an
.ethi cal judgment is in essence a statement about how—an
/"ideal observer" would react to a certain state of affairs or,

I variously, how any observer would react to a certain state of
vaffairs under "ideal" conditions. This is an exceedingly plaus-
ible reading of them. And although I now take a somewhat
different view of Hume's "intentions," I myself so construed
1 In Philosophy and Phenomenological Research, Vol. XIII, No. 3

(March 1952).
him in the introduction to my edition of Hume's moral and

political philosophy. 2 In that work I described Hume's posi-
tion, as it appears in the Treatise, in the following terms:
"(a) . Moral judgments are not merely expressions of
. .
our approval or disapproval; they have descriptive meaning

and are capable of truth and falsity; nevertheless, they are
empirical and hence corrigible, (b) Moral distinctions have }
a certain objectivity and universality in the sense that they re- /

fer to what an impartial and benevolent spectator would r ^
approve, and not necessarily to what most of us in fact do J
." 3
approve. . . What distinguishes Firth's analysis is the
care with which he analyzes characteristics of omniscienc e,
om nipercipie nc e, disinterestedne ss, di spassionaten ess, and
consistency which he takes to be the fundamental hall-
marks of the ideal spectator whose reactions we allegedly
have in mind in trying to decide what is morally right or
wrong.
Whether this particular list of characteristics which nota- —
bly does not include Hume's "benevolence" or any term usu-
ally associated with a particular moral attitude is really —
necessary and sufficient we may leave it to the absolute ob-
jectivists to debte a mong th emselves. But on any such list
the concept of gli sti^terestecl n^ss would doubtless be prom-
inently displayed. Moreover, Firth's analysis of it is typical of
his treatment of most of the others. For these reasons we may
treat his definition of disinterestedness as a test case for the
success or failure of his theory as a whole. If it can be
shown that any such definition, whose whole point must be

to eliminate all reference to sentiments or principles antece-
dently acknowledged as moral, to meet the test of
fails
plausible counter-instances, then, by a kind of intuitive in-

duction we may conclude that no supposedly formal prin-
ciple of objectivity can serve as a definitive principle of
moral right and wrong.
According to Firth's account, any observer is "disinter-
2 Hume's Moral and Political Philosophy, edited by H. D. Aiken

(New York: Hafner Publishing Company, 1948).
3 Ibid., p. xxxvi.
ested" and only if, he is totally lacking in "particular in-

if,
terests," where the phrase "particular interests" is taken to

refer to any interest whose object cannot be defined without
the use of proper names and such egocentric particulars as
"I," "here," "now," and "this."
4
Now it is no part of my pur-
pose to deny that such a definition might do for certain
senses of the term; the only question here is whether it
suffices for what we have in mind when we employ the term
in moral contexts. I magine, then, an observer who is devoid
of "particular" interests, in Firth's sense, but so single-minded
in his devotion to the pursuit of knowledge that he always
reacts with pleasure to any proposal that would increase
it, regardless of its consequences. Among such consequences,
we may easily suppose, would be a certain amount of un-

avoidable mutilation of young children, at least some confes-
sions obtained by torture, and a few instances of scientific
eavesdropping on persons at religious confession. Or, again,
imagine a great artist, also devoid of particular interests, who
is so dedicated to the art of painting that he always ignores
even the most elementary claims of veracity, kindness, or

loyalty when they interfere with the demands of the art. Now
I suppose that from a certain standpoint such individuals
may be regarded as "disinterested," perhaps even admira-
bly so. But if so, it is a form of disinterestedness which, as
such, has nothing to do with morals; on the contrary, it in-
volves forms of behavior which most of us would regard as
Drime examples of moral obtuseness, if not flagrant im-
\ morality. More important, we would probably also agree
y /that such individuals displayed a singular lack of the sort of
/disinterestedness required in moral situations.
But, for the sake of argument, let us assume that Firth's
definition of disinterestedness can, in principle, be amended
so as to rule out all such offending non-particular interests.
Accordingly, our formula will be that any person may be said
to be morally disinterested if, and only if (a) he is devoid
of particular interests ( in the sense defined above ) and ( b )
,
he is devoid of any non-particular interests of type G. But

4 Firth, op. cit.,
pp. 338-9.
how, pray, is this type to be defined? The suggestion which

may occur to some moralists is that it should include all
non-particular interests, professional, artistic, or institutional,*,

which would dispose one to be indifferent to the suffering J
of other persons. 5 But aside from the difficulty that the con-
'
cept of a person is itself by no means a neutral term in moral

contexts, it seems evident that we are now attempting to
write a particular moral principle into the very definition
of moral disinterestedness. In that case, however, the con-
cept of disinterestedness so far loses whatever virtue it may
have had which appeal could be
as a critical standard to
made in trying to decide, among other things, whether that
moral principle itself is objectively to be preferred to all
others, or whether, in certain circumstances, it should be
subject to exception.
The situationseems to be this: The concept of moral dis-
interestedness, and hence of moral objectivity, cannot be
successfully defined wholly without reference to prin-
ciples antecedently acknowledged as moral. But neither can
it be defined in terms of any particular moral principle
without at once losing its use as an independent standard

of objective moral appraisal. Firth's analysis avoids the
latter difficulty; its weakness is that it countenances forms of
"interest" which many moralists would regard as obviously
partial and which some would regard as immoral on the face
of it, whether partial or otherwise. Nor does there seem to
be any way of amending Firth's definition which would
remove that weakness without at the same time destroying
the very feature which originally appeared to commend it.
For the moment, however, let us waive the difficulty that
from a moral point of view many non-particular interests
seem, ethically, as partial as any particular ones. What we
have now to consider is the more delicate problem whether,
so far as morality as such is concerned, anyone saliently
affected by particular interests of any sort is to that extent
lacking in moral disinterestedness and objectivity. In brief,
must we suppose that a principled concern for "one's own"
5 Firth, I assume, would have no part of such a suggestion.
is always evidence of partiality and hence of a lack of ob-

jectivity? This does not appear to me at all obvious. Con-
sider, for example, a highly conscientious mother who, upon
careful reflection, always, and in good conscience, pre-
ferred the well-being of the members of her own family to
that of any other group in which they happened to be in-
cluded but which is definable in wholly non-particular terms.
Here it is mind that Firth's "particular in-
well to bear in
terests" are by no means always self-interested, and that a
person such as I have just described may be quite as selfless
as any Saint Just who ever dedicated himself to the "uni-
versal" principles of liberty, equality, and fraternity. Is such
a person, morally, any less disinterested than a Saint Just
or, better, a Tom Paine who so appealingly avows that
"Wherever there is injustice, there is my country?" On what
grounds that are not, at the same time, morally question-
begging? She is indifferent to "non-particular" claims that,
from the standpoint of her principles, appear impossibly
general and abstract; he is indifferent to "particular" claims
that, from her point of view, are always overriding. To him
she seems insufferably parochial; to her he seems virtually
inhuman, a monstrous justice-machine dedicated unfeelingly
to mere names of virtue. Which, morally, is the more disin-
terested? Without enlarging our own point of view there is,
so far as I can see, no way whatever to decide. One seems as
interested-disinterested as the other; the only difference be-
tween them is one of moral perspective. Once again, com-
pare a man who is exclusively dedicated to the welfare of
the members of his own country with another who is simi-
larly dedicated to the happiness of white protestants, what-
ever their country. Is the former person, because of his par-
ticular interest, any less impartial than the other? Without
regard to any other considerations already acknowledged
to be morally relevant, I simply do not see how it could be
denied that both parties are pretty much in the same boat.
Within the range of his principle, each may be completely
disinterested; with respect to his principle itself, one seems
quite as partial as the other.
It may be replied, of course, that a person may prefer his

own any "outsiders" without being guilty of any partial-
to
ity, but only on the condition that he is prepared to accept
the possibility that everyone else does so as well. In that case

his particular interest turns out, innocuously, to be merely
an application to his own situation of a non-particular con-
cern for a state of human affairs in which every individual
gives the nod to his own when their interests happen to con-
flict with those of outsiders. This reply is unsatisfactory. It
is not hard to imagine a person who would reply, upon re-
flection, thathe simply does not know whether to prefer a
world in which everyone gives the nod to his own, whoever
they are, but he knows, at any rate, that he ought to prefer
"his own" whenever their interests conflict with those of any
outside group which he can seriously envisage. Is his "re-
action" lacking in moral impartiality? But, again, on what
non-question-begging grounds? He has not refused the de-
mand to reconsider; he has entertained and then been
obliged to suspend judgment concerning the universalistic
covering principle in question. Despite this, his original
commitment has not been shaken. Once more, I contend,
there is no way of convicting him of partiality without intro-
ducing other moral principles which he himself has tem-
porarily forgotten, but which, upon further consideration,
he is prepared to acknowledge. Apart from them he is no
less (and no more) disinterested than any Crito whose dedi-
cated love of Socrates, as it turns out, is merely a disguised
interest in the wisest of all men. 6
From a morally disinterested point of view, the question is
not whether an interest is particular or otherwise, but only

whether, all things considered, it still appears right to re-
alize it. And among the things that saliently require to be
considered in such a case are precisely those prima facie
duties to which one is already conditionally committed. 7
6 Firth, op. cit., p. 339.

7 At this point it becomes desirable to remark upon the distinction
between lack of objectivity and mere subjectivity. Objectivity is a form
of achievement one can miss for a multitude of reasons; such reasons
Consider another problem which often presents itself to
the practicing moralist. From his own point of view he may

fail to be objective, not only because he is influenced by
extra-moral whether particular or general, but
interests,
also because of an obsession with moral rights of a particular
sort. Thus, for example, he may be overborne by a particular
prima facie duty to which he blindly refuses to acknowledge

exceptions he knows to be legitimate. Or he may give the
nod to a principle whose place in his moral scheme of things
is entirely subordinate, ignoring superordinate principles of
justice or humanity upon which the whole virtue of the prin-
ciple in question entirely depends. I am inclined to think
that failures moral disinterestedness occur most fre-
of
quently, not because we are overborne by extra-moral con-
siderations, but because we are temporarily blinded by par-
ticular moral sentiments, such as indignation, whose "rights"
are extremely limited. The would-be objective moralist is
thus always placed in double jeopardy; on the one side he
is tempted by a host of non-moral inclinations, both "partic-
ular" and general, and on the other by particular 8 duties that,

on second thought, have no such absolute claim to fulfillment
as he permits himself for the moment to believe.
In this connection it may also be well to mention a type of
have nothing necessarily to do with subjective preoccupations. On the

other hand, subjectivity itself may serve as a kind of principle which, in
certain contexts, functions as a test of one's powers of objective judg-
ment. In short, what causes a man to be non-objective in his decisions
and judgments is not simply a preoccupation with matters pertaining
essentially to himself, but, rather, those beliefs, attitudes, and interests,
whatever their objectives, that deflect him from the ends or prevent
him from conforming to the principles characteristic of the activity in
question. No doubt there are forms of activity with respect to which
any subjective concern is evidence of a lack of objectivity. But there are
others with respect to which lack of objectivity may be owing to per-
sistent intrusion of completely selfless general concerns that confuse or
corrupt one's sense of what is relevant to the form of activity in ques-
tion. Thus, for example, a Ricardian "economic man" loses objectivity,
not when he becomes completely preoccupied with his own greatest
material good, but when he allows sentimental "moral" considerations
to deflect him from the rational pursuit of that good itself.
8 Here the term "particular" is used in its ordinary, non-technical
moral problem which is frequently misconceived by moral

philosophers. Now it is true that in attempting to reach an
objective decision most moralists believe they have some ob-
ligation to listen to the considered opinions of other judges.
But this obligation, like any other, may itself become obses-
sive so that the moralist, ignoring other exigent demands
which his conscience makes upon him, loses objectivity
through the very attempt to achieve it. An individual may
well believe that he is morally obliged to suspend judgment,
for the time being at least, when he finds that there is a
strong consensus of moral opinion against him. But that ob-
ligation is one among many. And if he is overwhelmed by
it he may be as much incapacitated for objective judg-

ment as when he responds too promptly to his sense of in-
dignation. Contrary to the prevailing view in philosophy
there is, in general, no necessary comiection between ob-
jectivity of judgment and intersubjective agreement. Nor do
we automatically establish the objectivity of a judgment
merely by multiplying subjective opinions or attitudes. Inter-
subjective agreement, however extensive, remains nothing
more than that; it becomes a test of objective validity only
when, as in science, the very form of the activity demands
it. There are many forms of activity in which such agree-
ment is not required and where it exists does nothing to in-

crease the likelihood of objective certainty. In such cases,
too much concern for the opinions of others may well be the
greatest handicap to objective judgment. This is obviously
true in the case of literary and art criticism; it is true in the
domain of religious belief; it is true also in the sphere of
moral judgment where the demands of conscience may re-
quire one to stand completely alone.
What has been said above about the concept of moral dis-
interestedness applies, by analogy, to other "characteristics,"
such as dispassionateness, of an ideally objective observer.
Therefore, I shall not pause to make detailed criticisms of
Firth's treatment of them. It is time to make a more general
point. Now the common notion of moral objectivity (of
whose use I have already given some intimations) has no
—
tendency to breed a general "philosophical" skepticism about

the validity of moral judgments and principles. However,
Firth's absolute objectivism, when pressed, gradually forces
us back upon a principled moral agnosticism which, if taken
seriously, would result in the destruction of our ordinary
sense of right and wrong or, what comes to the same thing,
the complete paralysis of our moral will. "Judge not!" now
becomes not justan exemplary counsel of moral humility

but, in effect, the defining principle of morals itself. In a
word, if the traits of an "ideally" objective observer are taken
as providing the definitive principle of morals, then moral
right and wrong become vacuous conceptions, the use of
which answers to no conscientious human concern. This
point can be shown in a variety of ways. As we have al-
ready seen, Firth's analysis of disinterestedness fails to meet
the test of counter-instances. But the attempt to amend it
in a way which at the same time avoids reference to prin-
ciples antecedently acknowledged as moral, leads us first
into a blind search for a way to eliminate offending non-
particular interests, next to the admission that particular in-
terests are not necessarily less disinterested than any others,
until finally we begin to wonder whether all "interests"
including even the obligation to respect the judgments of
other persons —
must not be eliminated before we can secure
that pure and perfect disinterestedness which (by hypothe-
sis) is involved in the very meaning of moral right and
wrong. Disinterestedness thus becomes a kind of "hidden
God" which, just to the extent that the concept of morals is
made to depend upon it, forces us to question more and
more grimly whether that concept has any intelligible mean-
ing. In another way, the search for an ideal observer whose
attitudes and opinions are merely objective leads eventu-
ally to the conception of a being who has no favorable or
unfavorable reactions at all. We are reminded at this point
of Job, who, as he comes to realize that his notions of justice
provide no measure of God's justice, is forced to admit that
although he must be a sinner in God's eyes, he can never
know how or why. To this the reply must be that though
moral understanding is a pearl of great price it loses all value

if we make the price so high that it cannot possibly be paid.
THE NATURE OF MORAL OBJECTIVITY

The merit of the ideal spectator theory is that it enables us to
see just what absolute ethical objectivism comes to when we
try to formulate definitive principles of moral judgment, or
something called "the moral point of view," in total abstrac-
tion from any substantive principles and practices to which,
in the ordinary course, moral agents find themselves com-
mitted. Is such a point of view consistent with the ideal of
moral autonomy, to which, within limits, most moralists are
committed? It is impossible to say. Who knows how an ab-
solutely objective spectator, bereft of moral principles, would
in practice react to any situation? In effect, the spectator
theory would replace all of the little perplexities that beset
us when we worry about the objectivity of particular moral
judgments with one big perplexity about the nature and
reactions of the ideal spectator himself. Unhappily, the more
we contemplate this big perplexity the more we wonder
whether the idea of morality itself may not be a conceptual
monster, a mere idea of reason, to which no definite mean-
ing can be assigned. These doubts can be removed, not by
further logical analysis, but only by returning, conscien-
tiously, to the world of work-a-day moral problems and
judgments. In that world the problem of moral objectivity is
mainly a problem of piecemeal mutual adjustment of ac-
knowledged commitments within a loose framework of pre-
cepts and practices, none of which is ever permanently ear-
marked as an absolutely first principle and each of which is
subject to a list of exceptions that can never be exhuastively
stated. In practice, moral error is due not so much to our lack
of omniscience about matters of fact as to philosophical
misconceptions about the drift of the terms of our dis-
course and to our inveterate tendency to adhere inflexibly
to particular principles in the face of other loyalties whose
claims upon us are equally compelling. What is wanted is
not a better understanding of the hypothetical reactions of a

perfectly objective somebody-else, but that conscientious sec-
ond thought which enables us more general view of
to take a
our own existing responsibilities. Such a general view pro-
vides no definition of moral right and wrong; it does not re-
quire us to ignore "ourown" when we find that their claims
upon us cannot be universalized; it does not
conscientiously
demand that we treat "everybody," whoever they may be, as
moral persons; nor does it commit us to some supposed con-
sensus of moral opinion which all other "competent" moral
agents must be presumed to share. In brief, there is and can
be no absolute or universal vantage point from which con-
scientious moralists, regardless of their sentiments, may make
an objective appraisal of moral decision and
their particular
principles. Morally, we are always in the middle of things,
confronted with eternally exceptionable precepts which, un-
til such exceptions have been made, still lay presumptive
claims upon us that we cannot in conscience disavow. What

provides the basis for such exceptions? Nothing save other
particular principles which, in turn, we are forever driven to
qualify in the light of still other principles. And when we
come temporarily to the end of a line of qualifications what
then do we find? To our dismay, nothing but the very
"first-level" duties with which we began.
The ordinary principle of moral objectivity thus prescribes,
not that we
look beyond the moral life itself for a ground of
criticism, but only that we search within it for the soberest
and steadiest judgment of which, in the light of all relevant
obligations, we are capable. When a question arises con-
cerning the objectivity of a particular moral judgment or
principle, our task is always and only to look beyond it to the
other relevant commitments which we ourselves acknowl-
edge. And if this answer seems inadequate, then the reply
must be that there is, in conscience, nothing else to go on.
In the moral sphere it is always, finally, up to us; nor is there
anyone to whose steadier shoulders our burden of moral
judgment can be shifted. That is the agony of the moral life;
it is also its peculiar glory.
The principle of objectivity in morals is, then, essentially a

principle of reconsideration. What
demands, when a ques-
it
tion about the objectivity of a particular judgment or prin-

ciple arises, is that we consider whether such a judgment or
principle, as it stands, can be consistently upheld in the face
of whatever other moral considerations might be thought,
in conscience, to defeat it. What do such considerations in-
clude? It is beyond the scope of this essay to attempt a
codification, even if such were possible, of all of the sorts of
consideration that might, in principle, serve to defeat or fal-
sify a moral proposition. It must suffice here to indicate in
very general terms a few of the main factors involved, and
for the rest to remove certain prevailing misconceptions.
In many moral
situations the objectivity of a particular
judgment is sufficiently established simply by bringing it un-
der an appropriate covering principle which, in effect, sim-
ply classifies the action in question as an act of a certain sort,
the performance of which, other things equal, is a moral
duty. As a rule we determine at once that feature of an ac-
tion which enables us to classify it as an act of a certain
sort, the performance or non-performance of which is at once
required by a relevant covering principle. What creates a

problem for us are other features of the situation not covered
by the principle appealed to and which make of the action
in question something more, or less, than merely an act of a
certain sort. A certain line of action involves us, say, in the
act of telling a lie, and the telling of lies, other things equal,
is proscribed. But it may be that what is proposed, the action
to be performed, is more than an act of lying; perhaps it may
also be viewed as an act of kindness. In that case, it falls
under another principle which, other things equal, provides
the basis of a second obligation. In such a situation it is
commonly supposed either that there is a clear order of
—
precedence among such principles that is, an obvious prin-
—
ciple of hierarchy or else that both principles are merely
summary applications of some more fundamental principle,
such as the principle of utility, by which both such "first-
order" principles are justified and from which both are
ultimately derived. It is also commonly supposed that such

principles define certain distinct "practices," each of which
can be criticized only as a whole in the light of some
"second-order" principle, such as "justice" or "utility," which
alone provides an objective standard of appraisal. Such facile
descriptions of the ways out of our moral perplexities usually
will not do.
In the first place, moral principles cannot be arranged in a
flat hierarchical order. For example, we
cannot say without
qualification that the principle of kindness takes absolute
precedence over the principle of veracity; we cannot say
without qualification that promises may be broken when
the consequences of keeping a promise are more unpleasant
for the people involved than those of breaking it. It all de-
pends upon how much unpleasantness is involved in keeping
the promise or how much unkindness is involved in telling
the truth; conversely, itdepends upon our considered
also
views of the importance of telling the truth and of keeping
promises. For most of us, these practices are important com-
mitments in their own right which are not to be evaluated
simply by reckoning the amount of discomfort (or unkind-
ness) that keeping or breaking a promise, telling or not
/telling the truth, would involve. They are, in short, inde-
j
pendent sources of moral excellence which require no justi-
^-fication on merely utilitarian grounds. Secondly, it is a mistake
to suppose that what we call the practice of promising, for
example, can be understood in complete abstraction from the
principles of exception which, on occasion, justify the break-
ing of a promise. Part of what we understand by the practice
of promising is the recognition that promises may, under cer-
tain conditions, be broken and that no promise is absolutely
inviolate.Not only is every principle subject to exceptions,
but the fact that such exceptions are allowable belongs to
the very concept of a practice itself. In short, there is no such
thing as the practice of promising, the practice of veracity,
or the practice of justice independent of the principles of
exception which qualify such practices. This means that the
principles which form the parts of a moral code can be un-
derstood only in relation to a network of such principles, no

one of which can be evaluated in isolation from the rest. Nor
is there any unqualifiable "second-order" principle so abso-
lute and so impervious to exceptions that it provides a unique
and final basis for objective criticism of the rest. Objective
criticism of moral principle s, is mainly a matter of piece-
meal qualification of particular practices with a view tn th eir>
greater coherence within a moral s ystem. Thus the only test
of objectivity, so far as principles and practices are con-
cerned, is their ability to survive without further qualifica-
tion by the other principles or practices with which they
may conflict. When qualification is required, nothing pro-
vides an objective basis for decisions save the other practices
with which, by mutual adjustment, any particular practice
must be reconciled.
But there is a feature of the moral predicament still more
basic than any hitherto mentioned. Most philosophers hold
that what distinguishes moral from merely expedient action
is the fact that moral actions are performed not only on
the basis of, but also for the sake of, a principle. What is
right in one case is morally right only if it is so in cases of a
similar sort. Yet this approximate truth must accommodate
itself to another, no less fundamental. No action is exhaus-
tively definable simply as an act of a certain sort any more
than an individual substance is exhaustively definable as a
member of a certain species. It may exhibit morally signifi-
cant aspects beyond any that have been covered by our
prevailing system of moral acts. For this reason it is always
possible to maintain that a particular line of action ought to
be followed even though performance goes against every
its
principle in the book. In other terms, one may be more cer-

tain that a line of action ought to be carried out than of any
principle that might be thought to justify or to condemn it.
In that case, the action becomes, as it were, a principle unto
itself and at the same time establishes the basis for the in-
troduction of a new moral principle.

It is essential that I not be misunderstood in this. It is
theoretically possible to say of a particular moral judgment
that it is objectively defensible even though it cannot be

justified by appealany hitherto existing principles of jus-
to
But there is a world of difference be-
tification or exception.
tween a judgment which is sustained after a full and im-
partial review of the whole moral situation, including the
principles that appear to be applicable to it, and one for
which certainty is claimed in advance of inquiry. I am mak-
ing no defense of moral dogmatism; on the contrary, I main-
tain that the principle of objectivity may require reconsider-
ation of any judgment. But this applies no less to principles
themselves. It is quite as dogmatic and unobjective to main-
tain a moral principle regardless of the judgments which,
upon second thought, may disallow it as to insist upon a par-
ticular judgment without regard to the principles which
might be thought to invalidate it.
In summary, there is and can be no definitive criterion of
moral objectivity and, hence, no definitive principle of moral
right and wrong. When a serious question about the objectiv-
ity of a particular moral judgment or principle arises, there is
simply the further moral obligation to re-examine it in the
light of the other obligations and duties that have a bearing
upon it. If it should be replied that objective reconsidera-
tion requires, also, an endless search for new facts which, if
known, might alter our notions of our obligations and duties
themselves, the answer must be that such a search would
defeat the very purpose of moral reflection, which is judg-
ment. The principle of objectivity requires only that we take
account of any hitherto unconsidered facts to which we may
reasonably be expected to have access. But in that case, what
is reasonable? There is no formula for answering such a
question; our judgment can be formed only by weighing the

obligation to look for relevant facts against other obligations.
The principle of moral objectivity can neither supply the
materials for moral judgment nor tell us where to go in
search of them. If we have no time to search for further

possibly relevant facts, the principle of objectivity will pro-
vide us with not one moment more; if we are otherwise
lacking in moral sensibility, it will not make good our de-
—
ficiency by so much as a single obligation. What it can do
and it can do no more — is to dispose us to review our de-

cisions so that we may neglect no pertinent fact that is avail-
able to us in the time we have, and that we may neglect no
obligation which deserves be considered. Primarily, there-
to
fore, it what con-
functions as a principle of falsification, and
sistently survives the general scrutiny which it demands may
pass as objectively valid or true.
A NOTE ON MORAL TRUTH

This last remark provides, in essence, our final answer to
objectivism. But so profound are prevailing misconceptions
that a further word about moral truth is required before
we bring this essay to a close. Now
contemporary moral all
philosophers, regardless of school, take it for granted that

there is an intimate connection between the concepts of ob-
jectivity and truth. They may believe, as do the emotivists,
that neither concept has a moral application; or they may
hold, with the naturalists, that the conditions of moral ob-
jectivity and truth are no different from those of empirical
science. But they all agree that the connection between the
concepts is analytic. They all assume, moveover, that
truth, which they tend to restrict to statements verifiable by
the procedures of science or of formal logic, is logically the
more fundamental concept. On these points the following
statement of Professor Bernard Mayo is typical: "The deep-
est issue, and the most violent controversy, in contemporary
moral philosophy is between those who assert, and those
who deny, that moral judgments can be true or false. This
difference can indeed be taken as the simplest way of de-
fining the meaning of the pair of correlative terms 'ob-
jective' and 'subjective.' What is objective is capable of being
true or false, of being a statement, a belief or opinion; what
is subjective is not capable of a truth-value, but an ex-
is
pression of some psychological state." 9

Mayo's own view is
9 Bernard Mayo, Ethics and the Moral Life (New York: St Martin's
Press, 1958 ) p. 69.
,
that moral assertions, although subject to "criteria of cor-

by procedures akin to those em-
rectness" are not verifiable
ployed in science; he concludes, therefore, that they cannot
be true or false and hence that they cannot be regarded as
objective statements at all. 1 Professor Kurt Baier, who
takes virtually the same position as Mayo toward the rela-
tions of truth to objectivity, comes to precisely the opposite
conclusion. Accordingly, although he, like Mayo, holds that
moral judgments are essentially guides to conduct, he pulls
and hauls in order to show that by his definitions moral ap-
praisals are also empirically verifiable and hence that they
may be regarded as objective statements. 2 It seems not to
occur to either writer that verifiability (or falsifiability) is
not a concept for which empirical science alone has a use,
and that in certain domains questions of truth are settled
entirely by the conclusions of objective judgment itself.
Such at any rate, is the position of this essay: So far at
least as morals are concerned, the concept of objectivity is
logically more primitive than and the verifica-

that of truth,
tion of a moral proposition'bccurs when, and only when, we
judge objectively that it survives rescrutiny. The plausibility
of such an analysis is very great. We have already shown
that moral objectivity does not depend upon the possibility
that moral judgments are statements of fact, verifiable by
procedures analogous to those employed in empirical sci-
ence. We have shown, indeed, that moral objectivity does not
even presuppose that there are definitive principles of morals
of any sort. This, however, does not in the least imply that
1 Itshould be observed here that Mayo's analysis is inconsistent on

this point. Sincehe regards "objective" and "subjective" as correlative
terms, it would seem to follow that he regards ethical assertions, even
though subject to "criteria of correctness," as mere expressions of psy-
chological states. However, he goes so far at one point as to say that
"What we call objectivity in the former case [that is, of visual percep-
tion] we call impartiality in the latter [that is, the case of moral judg-
ment], and impartiality is what justifies us in saying that morality too
can be objective." Op. cit, p. 88. The incoherence of Mayo's analysis,
in fact, is beyond my powers to describe.
2 Kurt Baier, The Moral Point of View ( Ithaca, N.Y. Cornell Univer-
:
sity Press, 1958), p. 77 ff.

moral propositions are not statements. In its familiar applica-

tions, the term "statement" is by no means limited to empiri-
cal descriptions and predictions. Thus, even if it were
granted that only statements can be true or false, this leaves
us with a very wide range of possibly true or false utterances.
In addition to statements of ( empirical ) fact, there are state-
ments of intentions, statements of policy, statements of prin-
ciple, bank statements, and a hundred and one other forms
of statement, almost as various in their logical functions as
discourse do not, of course, claim that all statements
itself. I
are objective and true or false. I claim only that the use of the
concept of a statement is no more the exclusive preroga-
tive of the empirical scientist and the logician than is that
of objectivity, and that from the premise that a statem ent
is non-factual, in the empirical s ense, nothing whatevgr^can
be inferiedT about the possibility of Its being verifiable. The

only fundamental question, so far as verifiability is con-
cerned, is whether the statement in question is corrigible,
and whether, when we affirm it, we are in any meaningful
sense subject to correction. In the case of morals, I contend
that possibility is guaranteed by (and only by) the possi-
bility of objective judgment itself. 3
To say, then, that a moral statement is true is simply: (a)"
to r eaffirm it and (b) to avow that it meets whatever tests o f
objectivity are deemed propeT~rJ\nffie~moralJ udge himself.
He who affirms that a moral proposition is true when it will
not abide such tests, speaks falsely. No moral judge, in
affirming the truth of a particular judgment or principle, sup-
poses that anyone else must agree with him regardless of
his own moral obligations. For in morals there can be no
guarantee that all objective judges will acknowledge the
same principles of moral obligation.
This means that rational disagreements in morals cannot
range beyond an implicit framework of common principles
and practices shared, in conscience, by the disagreeing
parties. Similarly _jmeanmgful interpersonal discussions of
J!
3 I believe that the situation is much the same in many other spheres
of discourse. However, I am arguing here only the case of moral truth.
'th e (CTuth^f a particular moral judgment presuppose the

existence of a moral community to wh ichJa^nnsoi Ftriop: both
discussan ts are committed. This does not preclude the possi-
bility that moral judges may contradict one another; it pre-
cludes only the possibility that contradiction can occur out-
side the limits of a particular moral community. To the
extent that moral communities differ fundamentally in regard
to their moral precepts and practices, it is pointless to speak
of their judgments as logically contradictory. For in that case
it is also pointless to talk of an objective settlement of their
differences. At such a what is wanted is not argu-
point,
ment, but education; not the appeal to non-existent princi-
ples of morals, but companionship and love.
[IX]
God and Evil:
A Study of Some Relations
Between Faith and Morals
THE CONTEXT OF THE PROBLEM OF EVIL

To outsiders, there has always been a grim fascination in the
fact that Western monotheism, which arrogates to its God
exclusive power to redeem us from evil, should itself be so
bedeviled by the apparently simple yet evident proposition
that something is evil. But even such comparative insiders as
Augustine and Dostoevski have been obsessed by the irony
implicit in the fact that, having hopefully begun by justifying
themselves to God, they must end despairingly by having to
find excuses for him. j Between man's justice and God's Jus-
tice, as Abraham and Job discovered, there falls a shadow
whose darkness becomes only the more inscrutable as the

light of human reason is trained upon it. This enigma, or
paradox, which marks the profoundest crisis of faith and
morals that the monotheist knows, is usually referred to as
"the problem of evil."\ It might, with more propriety, be
labeled "the problem of faith and conscience." Compared
with it, the metaphysical perplexities inherent in the notions
of a causa sui or a necessary being are, in Locke's sense,
trifling. The latter belong to an episode in the history of a
rationalistic theology whose connection with the faith of the
prophets, the apostles, and the fathers of the church remains

problematic^ The problem of evil is of a different order. What
it concerns is the ultimate consistency of a form of life, and
what it threatens is the faith animating that form of life.\No
solution to it would extend by a hair's breadth either our
grasp of relations of ideas or our knowledge concerning mat-
ters of fact. It would be more appropriate, indeed, to speak,
not of solutions to the problem of evil, but of modes of abso-
lution from it.
In these gay and gaudy days of existentialism and neo-

orthodoxy, such remarks may seem merely to underline
the dangers of consecutive thinking about religious and
ethical problems. Such was not their intention. A problem
does not transcend logic merely because we can call it
practical or existential. The inconsistency between the faith

and the conscience of the traditional monotheist cannot be
removed, certainly, by the methods available to empirical
science, mathematics, or linguistic philosophy. But the very
fact that it can be stated with some precision ought to be
sufficient evidence that religious utterances, as well as moral
judgments, occasionally have ponderable implications. And
our recognition that it is an inconsistency is at the same time
an implicit acknowledgment of the rational demand that it
be removed.
There is nothing very devastating in these truisms save for
those who are tempted to protect their faith by talking non-
sense about their God. "If you make up a self-contradictory
sentence," Professor Anthony Flew has observed, "it won't
miraculously become sense just because you have put
the word 'God' as its subject." It should be added that ob-
scurity and obfuscation do not suddenly become edifying
when transposed into a theological key. Opinions may differ
as to the regenerative value of continual preoccupation with
matters of ultimate concern. But there should be no room for
doubt that ultimate concerns which cannot be consistently
formulated are offenses against reason and hence offenses
against the human spirit itself.
A more serious question is whether, within the existential

God and Evil 173
context of faith, the problem of evil may not be unreal. It has

been said that the question of whether God exists has no
meaning for the unconverted, but that for the converted it
does not arise. It has likewise been argued that the ques-
tion of whether God's will is good has no significance, since
for men of faith his will is, as we say, "good by definition,"
whereas for unbelievers, who deny that there is a God,
any question concerning the merits of his will is meaningless.
Within the context of religious faith, which is the only con-
text where such matters could be discussed, there is, in brief,
no such thing as the problem of evil. Outside that context it
is only a formal dilemma which has no interest save as an
exercise in elementary logic.

\ Before replying to this objection, worth remarking
it is
that it is who have posed

not merely atheists and skeptics
the problem of evil. It is already present in Job, and it is ex-
pressed with great eloquence by Augustine and by Dostoev-
ski. Such an ad hominem is, however, formally question-
|
begging, and, formally, I stake nothing on it. It is more

germane to inquire whether conversion has any meaning
apart from the possibility of apostasy or, as I say, "decon-
version." (The proper word, "perversion," has lost its use-
fulness in this connection. ) It is also necessary to point out
in this connection that what we call "faith" is, in most in-
stances, merely an aspiration in that direction. If "outsiders"
cannot meaningfully question the existence or the goodness
of God, then, since all of us are virtual outsiders, who is in
a position to assert this? Limiting questions occur to every
man, and every man, "meaningfully" or not, has doubts
about the goodness of God and, through them, about His
existence. Existentially, deconversion, or loss of faith, is a
constant threat which any person can, at best, only hope
to overcome. Even those like Nietzsche, who announce that
"God is dead," may be said to know whereof they speak.
Animals may lack the power to understand religious con-
cepts but, if so, this is not owing to any lack of faith on their
part. On the contrary, lack of faith is, logically, as sound evi-
dence of understanding as faith itself.
We tend, I think, to confuse the issue by not bearing in

mind the distinction between faith and creed. No problem
about God's goodness or existence is explicitly acknowledged
by those entrusted with the formulation of a creed. Why
should be? It does not follow from this, however, that
it
such problems have no meaning for those who profess the

creed. For a Dostoevski, the problem of evil is not so much
the problem of Christianity as it is his problem. The creed is
a formula which is acceptable or not, depending upon his
ability to satisfy the joint demands of faith, conscience, and
reason. These are not, as it were, the demands of self-
activating and dissociated spheres of which are im-
activity
pervious to what transpires in the rest of They are the life.
interlocking demands of individual men or groups of men.

There is a sense in which we may perhaps agree that the
heart has reasons which the intellect knows not of; but they
are reasons nonetheless, and what does not satisfy the hearts
(and minds) of many men is the possibility that anything
can be evil if there is an almighty God who is also per-
fectly good.
A STATEMENT OF THE PROBLEM

lit is not my intention in this essay to offer any particular
form of relief from the problem of evil. My present interest in

the problem is philosophical :\ I desire merely to consider
what sort of problem it is and to determine the significance
of the various theological or ethical moves that have been
made in coping with it. This investigation belongs less to
I
theology, which is itself a form of religious discourse, than to

—
the philosophy of religion and of morals to the logical anal-
ysis, that is, of religious and moral utterances and of the
forms of life of which they are at once a manifestation and

a part.
There are many ways of stating the problem of evil, de-
pending upon the precise nature of the faith in question and
the particular evils, natural or moral, whose existence is ac-
knowledged. Because of its comparative brevity, it will be
God and Evil 175
convenient to present here a modified restatement of the

problem it in Part X of Hume's Dia-
as Philo formulates
logues. It be understood that no claims are made here
is to
for the merits of the theology involved; from this point of
view, my exposition will be as neutral as I can make it. In
short, I adopt this formulation for purposes of analysis and
only, therefore, as a working model. Many refinements
might easily be introduced into it, but I am convinced that
they would merely complicate our discussion without in-
creasing our understanding of the concepts and the propo-
sitions involved.
As here understood, the problem of evil tends to arise
whenever there is a disposition on the part of anyone to as-
sent to both of the following propositions: (a) there is an
almighty and omniscient being who is a perfectly good
person and who alone is God; and (
b ) there is something in
the finite universe created by that being which is evil. For
I
purposes of discussion, they may be referred to respectively

as (a) the theological thesis, and (b) the ethical thesis. iTheir
inconsistency may be By hypothe-
seen in the following way:
sis, an almighty and omniscient being can do whatever it
wills. But any perfectly good person, so far as he can, will do
good and prevent evil. Now, if there is a being that is at
once almighty, omniscient, and perfectly good, it will be
both able and willing to prevent evil. On the other hand, if
something is evil, it must be concluded that there is no such
being, since a perfectly good person would prevent it if he
could, and an almighty and omniscient being could prevent
it if it would. Either, then, there is no such being or nothing
is evil. But since, by hypothesis, only such a being is God,
we are forced to conclude either that there is no God or else

that there is nothing which is evil. |
This dilemma, however, is a serious problem only for the
believer who accepts or aspires to accept both the theologi-
cal and the ethical theses. His predicament may be expressed
in the following way: Both propositions seem to him to be
certain; the theological thesis is certified by faith, the ethical
thesis by conscience. If he abides by the deliverances of
conscience his faith is imperiled; but if he accepts the faith
to which he aspires, he must evidently disavow what, in

conscience, he knows to be evil. How can he deny what
his conscience tells him to be evil? Yet how can he deny
what by faith he believes to be God? Does his conscience
delude him, or is his faith somehow a mistaken or even a
wicked faith? What can he do? In principle, he may modify
his conception of God; he may change his attitude toward
what he has considered to be evil; he may try to find excuses
for God under the auspices of theodicy; lor, finally, he may
try to live with his contradiction as an honest man who de-
spairs of his soul's integrity.Whatever he does, his way of
life will never be quite the same again, for something that
has hitherto been a cornerstone of that way of life has now

been shaken to its foundations.
THE MEANING OF THE THEOLOGICAL THESIS

Before examining the moves that may be made in order to
relieve the problem of be necessary to give some
evil, it will
attention to the theses themselves. For our understanding of

the problem and the moves themselves will depend en-
tirely upon what we take the propositions that give rise to it
to mean.
The theological thesis has several parts which are not al-
ways clearly distinguished. It is not to be supposed that,
taken as a whole, the thesis explicates the meaning of "God,"
even for the monotheist himself. The fundamental reason
why God has no essence is that no essence can exhaust the
idea of the holy or the worshipful. In theology, as in ethics,
definitions are commonly misleading. Where they occur, they
serve not as explications of the use of expressions, but as
devices for the focusing and directing of sentiments or atti-
tudes. They explain nothing, but signify much, and what

they signify is not something about the nature of things in
themselves but rather the drift of our own sense of what is
holy or worshipful. It is in this way, I take it, that we must

understand how such theologians as Aquinas can argue that
God and Evil 177
God has no essence, while at the same time discoursing at

length about his inherent power, personality, and goodness.
The first part of the theological thesis may be called the
metaphysical claim. What it asserts isis an al-
that there
mighty and omniscient being. Now the metaphysical attri-
butes of omniscience and omnipotence are not, of course,
entirely free from internal difficulties of their own. It has been
argued that no consistent idea of them can be formed at all,
and that when we stretch the notions of potency or power
and knowledge to infinity and undertake to conceive them
in absolute or unconditional terms we always land in
paradox and confusion. I myself am inclined to think that
any terms which have a persistent and common use also have
or gradually acquire a meaning, and that if we bear in
mind the context in which they are characteristically em-
ployed, we will usually find an idea which is neither vacuous
nor inconsistent. I intend this,however, less as a "theory"
than as a heuristic principle of interpretation to which we
are free to admit exceptions in particular cases.
As R. B. Perry has remarked:
God conceived as perfection reflects man's experience of im-

perfection. Practicing and suffering injustice, man dreams of
perfect justice; being both hateful and the victim of hate, he
dreams of utter and universal love; being ignorant, he conceives
of omniscience; amidst the ugliness and drabness of life, he fan-
cies a perfection of beauty; from his unhappiness there springs
a vision of perfect and uninterrupted bliss. 1
The point is that "omniscience" primarily represents the

overcoming of our own deprivation of knowledge or, rather,
the ideal overcoming of this deprivation. As such it signifies
an ideal of knowledgeability to which we ourselves con-
ceivably may aspire and which we may endlessly approach.
Similarly, "omnipotence" represents an ideal of power or
puissance, of ability to do and to accomplish whatever
one may will, without external interferences of any kind.
1 Ralph Barton Perry, Realms of Value (Cambridge: Harvard Uni-
versity Press, 1954), P- 483-
Both characteristics are ascribed to God as inverse symbols

and measures of our own imperfections and limitations.
Some hold that the metaphysical attributes must be
ascribed to God in a merely analogical sense, on the ground
that God is a being so utterly unlike ourselves. To me this
seems problematic, since God, conceived as a "Thou," can-
not be absolutely unlike ourselves. Such attributions merely
tend to be regarded as analogical when we re-emphasize
the distance between ourselves and God. Briefly, whether
such attributions are to be regarded as analogical or not de-
pends precisely upon how we ourselves conceive of God,
upon how near or far, like or unlike, he appears to us to be.
But this will depend, in every case, upon the particular
stresses to which the religious life itself happens to be sub-
ject. It is my contention that one primary motive for insisting
upon the "merely" analogical or even metaphorical character

of all such attributions as "omnipotence" to God is the felt
necessity to overcome the inconsistencies involved in the
problem of evil. For in regarding his "power" as merely an-
alogical or metaphorical we seem, at least, to relieve our-
selves of the necessity of treating him as an agent, like our-
selves, who does things and who is therefore responsible, in
the ordinary way, for what he does.
Be may, God is presented, throughout the biblical
this as it
literature, as one who acts and as one who is only in the
manner of an active being. Might is ascribed to him as one
who performs prodigious acts and who does deeds of ines-
timable importance for mankind. Might itself is essentially
the power to do something, and someone is mighty in so
far as he has the power to do and to act. Thus, in speaking of
God as "almighty," one is saying that the object of worship,
the holy one, possesses unlimited might or power, that it has
power over all, especially over ourselves, and that nothing
has power over it. And in asserting that there is an almighty
or omnipotent being one is saying that there is something
whose nature it is to act and whose power is unrestricted.
As for the attribute of omniscience, it is surely a radical
error to restrict its meaning within the religious context in
God and Evil 179
accordance with the preconceptions of some special theory

of knowledge such as empiricism or rationalism. The term
had a use before philosophers undertook to solve their
epistemological puzzles and before they attempted to estab-
lish specific criteria of knowability. So far as religion is con-
cerned, I think it suffices to say that an omniscient being
would know whatever there is to know, and that just as
Omnipotence is not expected by sensible men to do the
impossible, so Omniscience is not expected to know more
about the impossible than that it is such. What is logically
unknowable could not be known even by an omniscient be-
ing. Such a being would not know how to square the
circle; on the contrary, it would know with blinding clarity
that the circle cannot be squared. Hence, if anyone should
claim that an omniscient being would understand how in-
justice can be just or how evil can be good our best recourse
would be simply to stare him out of countenance. Should a
question be raised as to God's ability to know everything that
will happen in the future, we have only to consider whether
anyone can properly be said to know that something will
happen in the future. An omniscient being will know what-
ever there is to know about the future; and if it should
be that the future cannot be known, then not even an
omniscient being could be supposed to know it. How omni-
science might be come by is an interesting question, but I
think we do not have to consider it. It is hardly for us to ask
how the deity may have got into his line of work. Nor need
we consider here whether there is an omniscient and om-
nipotent being. For in dealing with the problem of evil, we
are concerned with a situation in which certain propositions
have already been premissed and which, for those involved in
it, are accepted on faith.
THE MORAL CLAIM OF THE THEOLOGICAL THESIS

The second part of the theological thesis may be called the
moral claim. What it asserts is that there is an almighty and
omniscient being which is also a perfectly good person.
More specifically, it asserts that such a being is both a per-

son and good — good in the way that a person is so. For
present purposes, it be unnecessary to consider the

will
many other ways in which God may be good, since they are
not directly involved in the problem of evil. Many people, no
doubt, will wish to say that God is good not only in the way
that persons may be good but also in the way that beautiful
things are good. There are conceptions of deity, such as
that of Plotinus, which are more aesthetic than moral. But
traditional theism, while it has its aesthetic aspect, is more
fundamentally concerned with God's justice and benevolence
than with his radiance.
Now, there is no logical connection between the meta-
physical attributes and the moral attributes. Logically, there
is no reason why an almighty and omniscient being might
not be a perfect stinker. Nor is it logically impossible that
such a being might be worshipped as God, even if it were
denied the attributes of a good person. The notion of a God
of pure power is not attractive, but it is conceivable. But
from Genesis on, the goodness of the almighty being is con-
stantly reiterated, and even the imponderable "I am that I
am" plainly implies by the presence of the first-person pro-
noun that whatever else it is, "the being" is not to be con-
ceived as a mere thing. It is not difficult to attenuate God's
personality, and those who would relieve him of responsibil-
ity for evil are likely tobe attracted to this way out of their
perplexities. But they do so at a price which is greater than
traditional monotheism has usually been willing to pay.
The term "good," as everyone now knows, is a general
term of commendation and praise. In using it, we normally
perform what Professor John Austin calls an "illocutionary
act." That is, in saying that something is good, we are doing
something with words, and in this case what we are doing is
to praise or to commend the thing in question in some way.
Ordinary commendations, of course, are always qualified
and, in context, they presuppose some standard in relation
to which the commendation is made. We have here to do
only with moral goodness, and hence with moral praise or
God and Evil 181
commendation, for we are considering God only as one who

acts and whose merits are those of a good person.
The notion of personality, what it is to be a person, is more
elusive and more ambiguous than that of goodness, and con-
temporary philosophers have devoted far less attention to
its analysis. We are not here concerned with "personality" in
the sense in which that term is used by empirical psycholo-

gists, such as Professor Allport, when they talk about the
"psychology of personality" —the psychology, that is, of in-

dividual character and temperament. As employed in ethics
and in theology, personality is not an empirical characteris-
tic like green or hot; nor is it exhaustively definable in terms
of empirical characteristics. Being a person, like being a
work of art, is which has to do with a
a functional conception
characteristic way of treating or dealing with the thing in
question. This fact is most easily seen in the following way:
It is always proper to speak of "treating a person as such" or
of "treating a work of art as such," and there is always a
point, at once understood, in saying that something should
be treated as a work of art or that someone should be
treated as a person. But were it asserted that a mountain,
a thunderstorm, or a green object should be treated as such,
or if someone should speak of "treating a mountain as such,"
the point of the remark would not be readily understood and
we would be obliged to ask the speaker what in the world
he meant by it. It is not that mountains cannot be treated
in various ways. On the contrary, that is just the trouble:
mountains can be treated or dealt with in innumerable ways
according to our own interest or pleasure. But just because
of this there is no way of treating them as such, no character-
istic way that the being of a mountain as such requires.
In the sense in question, then, ascriptions of personality

differ radicallyfrom descriptions of the empirical characteris-
tics of things. In speaking ofsomeone as a person, we are
assigning to him a certain function, role, or status and con-
ferring upon him the title of personality which goes with that
function, role, or status. And in addressing someone as a
person, we are thereby dealing with him in that role, and are
entering into those relations with him that such a role may
prescribe. In this sense, a personal relationship consists, not
in the innumerable ad hoc dealings which individual persons
happen to have with one another, but, rather, in the re-
stricted class of dealings that are prescribed by the role and
which may affect the status of the individuals involved as
persons.
A personal relationship may be properly said to fail when
the parties to it are correctly charged with failing in their
mutual responsibilities as persons. If it fails absolutely, then,
so far at least as the individuals themselves are concerned,
their very existence as persons is placed in jeopardy. As in
other spheres, repudiation of the role means abandoning the
go with the role, and vice versa. It is thus
responsibilities that
no whether someone is a person or whether we
idle question
ourselves stand in a personal relationship to him. For upon
the answer to it depends the validity of the ascription to him
of a whole system of rights and responsibilities. The denial of
a personal relationship does not involve the cessation of
all dealings on the part of the individuals involved. But it
amounts to the active repudiation of specific commitments
to deal and to be dealt with in certain characteristic ways. In
short, there is already built into the notion of personality a
characteristic normative function which prescribes a certain
comportment toward those to whom personality is ascribed
and which at the same time imputes to them certain inescap-
able liabilities and responsibilities.
There are, of course, many kinds of personality, each carry-
ing with it a distinctive role and status. Juridical persons are
thus distinguishable from moral persons essentially by the
different prerogatives and liabilities that go with their re-
spective roles. The rights and responsibilities of juridical
persons are defined by law, and may be altered as the law
requires. The rights and responsibilities associated with
moral persons pertain to moral agency as such, and they will
change in accordance with modifications in our conception
of the moral agent. In our system they include, above all, the
right to independent judgment and choice, and the reciprocal
God and Evil 183
and to choose in ac-

responsibility to judge disinterestedly
cordance with principles of right action that are equally
binding upon all persons.
Every moral person is liable to moral praise and blame.
In fact, anyone for whom moral personality is claimed is ipso
facto subject to moral praise and blame as a "good" or "bad"

person, since such praise or blame is equivalent to the as-
sertion that he has fulfilled or failed in his responsibilities as
a person. Anyone who acquits himself of his moral respon-
sibilities thereby deserves to be praised as a good person,
and anyone who fails to do so thereby deserves to be cen-

sured as a bad person. Should it be argued that a certain
person is not, in principle, subject to moral criticism, this
would amount to saying that he cannot be charged with
moral responsibilities. By the same token, to remove an in-
dividual altogether from the sphere of possible moral rep-
robation is so far to cease to regard him as a moral agent
and hence as a moral personality.
If this is so, it would appear to be impossible to speak of
an almighty and omniscient being as a good person without in
some degree treating it as a moral agent and hence as, in
principle, subject to the blame to which moral agents are
liable. And to the extent that we address such a being as a
person or as a "thou" we automatically treat it as an agent
to whom certain obligations are due and from whom the
fulfilment of corresponding obligations may be expected.
There may be reason to deny that such a being can reason-
ably be regarded in this way. Thus it might be argued that it
would be pointless to address an almighty being as one
would address ordinary mortals who are subject to our cen-
sure as moral agents. How, it may be asked, can we reason-
ably think of such a being as beholden to us in any way? It is
not my purpose here to assess the merits of such arguments
and questions. I contend only that if, for whatever reason,
we should seek to place such a being altogether beyond
moral good and evil, the effect would be, so far as we our-
selves are concerned, to relieve it of certain attributes of
personality and so to make it impossible to address the be-
ing,with a straight face, as a "thou." Hence, just to the ex-

tent thatwe are impelled to say that the "justice" of such a
being has nothing to do with our standards of justice or
are moved to stress the "mere" analogy between its "good-
ness" and that which we ascribe to ordinary mortals when
we them as moral agents, we thereby radically at-
consider
tenuate its status as a person and jeopardize any moral claims
that it might be alleged to impose upon ourselves. On the
other hand, we may speak, if we will or must, of "the al-
mighty to whom all praise is due"; but if we do so in a sense
that has any ethical significance, we render it possible also
to speak of "the almighty to whom all blame is due."
A "holy person" does not mean the same thing as a "moral
person." A holy person is onewho is entitled, as such, to
religious devotion or regard, and for whom the status of
holiness is claimed. A moral person becomes a religious per-
son and the moral "thou" becomes a religious "Thou" just to
the extent that the former is regarded as something which is
to be treated reverentially — just to the extent, that is to say,

that his principles, his conduct, and his volitions are not
merely praiseworthy, in the moral sense, but also holy or
divine. Certain persons, such as Jesus, are thought to be
morally exemplary by many who would not dream of regard-
ing them as divine.
On the other hand, it is conceivable that something might
be treated as a holy person without at the same time being
regarded as a good person in the moral sense. It is also
logically possible that ascriptions of moral personality might
be defeated without thereby entailing the defeat of ascrip-
tions of holiness or divinity. In the case of ethical religions
such as Western monotheism the relations of religious person-
ality tomoral personality are far from simple. A certain ten-
sion between the holy personality and the moral personality
of God is bound to occur in so far as one is obliged to treat,
as holy, principles or actions which, on their own account,
may appear to be reprehensible. No doubt the moral per-
sonality of God is to be conceived as deriving from or as de-
pendent upon his holy personality. But because of this, the
God and Evil 185
holy personality of God is placed in jeopardy when his per-

fection as a moral person is impugned.
It is therefore entirely understandable that certain mono-
theistic theologians should have sought to treat the holiness
of almighty God as something which absolutely transcends his
moral personality. Yet a too emphatic emphasis upon the
absolute transcendence of God's holy self is bound to result
in a tendency toward the secularization of the moral and
hence in a radical circumscription of the domain of the
religious itself. Such tendencies are plainly discernible in
Kierkegaard and even, at times, in Martin Buber. On the
other hand, identification of God's holiness with his justice
and his law may result in the defeat of the faith in him as
a divine being, when the validity of his justice and his law
has been placed in question. If his divinity may provide the
guaranty for his wisdom, so, in turn, the authenticity of his
wisdom tends to become a criterion of his divinity. This
hazard is inherent in every ethical religion, and no amount of
purely dialectical maneuvering will permanently remove it.
THE RELIGIOUS CLAIM OF THE THEOLOGICAL THESIS

The most is that which
crucial part of the theological thesis
asserts that the one almighty, omniscient, and perfectly good
person is God. I shall call this the religious claim. The most
important thing to be said about it is that the word "God"
functions at once as a holy name and as a title which serves
to deify that upon which it is conferred. The active verb
"to deify" is illuminating in this connection precisely because
it brings out the point that in calling something "God" we
are giving it a holy name and conferring upon it the status of

divinity. Or, to put the matter in another way, in speaking of
something as "God" we are using a name which at the same
time invests the being with the prerogatives, responsibilities,
and claims upon ourselves to which that name entitles it.
The apparent circularity in this analysis is removed when
it is observed that in deifying something we are treating it
as an object which is holy, worthy of our worship and

reverence. If any beingis God, then by definition it is worthy
to be worshipped and ought to be worshipped. Because of

this, those who do not worship that being which is God are
ipso facto reprehensible or blameworthy. It is also for this

reason that atheism is not generally regarded by the pious as
one philosophical or theological position among many, but

as a wicked or sinful attitude. Atheism is not the same thing as
the denial that there is an almighty being. One is an atheist,
in the strict sense, only if he asserts that there is no God,
which is to say, no being that is holy or worthy of worship.
In order to reinforce the validity of this analysis, let me call
attention to the essential difference between the meaning of
"God" and that of "god." The word "god" is not at all a holy
name, and its application to any being carries with it no
titular claims whatever. It may be used by anyone without
the slightest suggestion of blasphemy or impiety. It is, in fact,
impossible to take the name "god" in vain, for it is not a
name at all, but, like "cat" or "tree," a mere common noun.
There is nothing blameworthy in denying the existence of a
god or in refusing to worship a god, for there is nothing
intrinsically worshipful about the being of a god.
THE MONOTHEISTIC "SYNDROME"

We have now to consider the theological thesis as a whole.
If I am is no logical contradiction in accepting
right, there
any one of the three main claims embodied in the thesis
while denying the rest. One may accept the metaphysical
claim without accepting the moral or the religious claim;
and one may accept the metaphysical and moral claims with-
out necessarily accepting the religious claim. In principle, it
would also be possible to accept the purely religious claim

that God exists or that there is a being that is God without
thereby accepting either the metaphysical or the moral
claims. In brief, it is not logically necessary that God should
be almighty and omniscient or even a perfectly good person,
and it is not necessary that an almighty being or even a per-
fectly good person should be God. For the ordinary mono-
God and Evil 187
theist, nevertheless, such logical possibilities are not theo-

logically conceivable, or at any rate they seem not to be so
until the problem of evil is confronted. So regarded, the
theological thesis may be called "the monotheistic syndrome."
Given this syndrome, God "must" be viewed as an almighty,
omniscient, and perfectly good person, and conversely, if
there isan almighty, omniscient, and perfectly good person,

that person is alone entitled to be called God.
What, logically, is the import of this syndrome? It is cer-
tainly not an analytic truth. Nor is it as such an ordinary
statement of fact. It is more illuminating to compare its
function with that of moral principles. Now, as everyone
knows, it is always logically possible to question whether
happiness and pleasure are good. The ordinary usage of
"good" allows for the possibility of such a question, and those
who speak of "good" and "happiness" or "pleasure" as syno-
nyms are simply mistaken. For the ethical hedonist, however,
it is, in effect, not morally possible to ask whether pleasure
alone is good. For him, although "pleasure" and "good" do not

mean the same thing, pleasure is still held to be the only
intrinsic good, the only thing, therefore, which ought to be
desired for its own sake.
It is not very helpful to speak of ethical principles as a
priori synthetic. Were we so to speak of them, it would be
only in order to emphasize the point that it is a mistake to
treat them either as tautologies or as contingent ascriptions
of value. For those who accept it, the greatest-happiness
principle is not at all analytic; nor could its validity be es-
tablished by inspection of the meaning of the phrases in
question. Taken as a principle, it serves, rather, to prescribe
the standard of right action which is to be used in justifying
particular maxims of conduct. When it is accepted, the con-
tingency that a particular pleasure may not be intrinsically
good is excluded, as we
on principle, and the moral pos-
say,
sibility that a right action would not conduce to the greatest
happiness is precluded a priori from one's moral reckonings.
Such a conception of moral principles provides a useful
analogy, I think, for interpreting the monotheistic syndrome.
For the monotheist it functions, in effect, as a religious prin-

ciple which pre-empts the holy name and title of "God"
for the almighty, omniscient, and perfect being. I think this
explains the grain of truth implicit in R. M. Hare's view
that theological theses are not falsifiable. To put it bluntly,
they are not falsifiable just because they are so treated;
because, that is to say, the person who accepts them refuses,
on principle, to consider seriously any other alternative and
because he refuses to allow for the contingency that they
might turn out to be false.
But now I shall appear to contradict myself. For I wish to
say that while the convinced monotheist regards the theo-
logical thesis as a principlewhich is not falsifiable, still it is
not, in practice, absolutely beyond question, even for him.
Here, I confess, I find myself in need of words. But it is
necessary to press the point, even at the risk of apparent
contradiction, for it takes us, as I believe, to the very heart
of the spiritual perplexities with which we here have to
do.
It is significant, in this connection, that we do not so nat-
urally say that a theological thesis is proved or disproved as
that particular individuals have faith or lose it, and are sub-
ject to conversion and apostasy. And it is relative to our
faith or lack of it, our conversion to or deconversion from a
theological thesis, that we may say that the thesis is, in some
sense, religiously defeasible. At any rate, in speaking here of
the defeasibility of a theological thesis, I will mean that with
respect to it we are subject to conversion or deconversion,
and that our faith in it, despite the fact that we may hold it
beyond all doubt, may in fact be defeated by considerations
which, as Mill would say, are capable of influencing the
mind.
sometimes thought that faith and conversion, while
It is
subject to causes, are impervious to reasons; and that while

they may have, so to say, a psycho-logic they have no logic.
This is simply not true. Pascal, the arch-fideist, claimed that
the heart has reasons that the intellect knows not of, but he
still claimed, significantly, that the heart has reasons, not that
God and Evil 189
it is subject to causes. I should prefer to say that faith, al-

though it may perhaps have no reasons, is nevertheless sub-
ject to reasons, and that it may be weakened or else defeated
by reasons, as well as strengthened or confirmed by them. I
think it is in this light that the proofs for the existence of
God are to be understood from a religious point of view:
they are not really proofs at all but, for those who accept
them, confirming evidences of a faith which might still be sus-

tained without them.
Although, despite Spinoza, there are thus no Q.E.D's in
the domain of religion, there still are reasons, and reasons
may as well prepare the way to a conversion as may a mystic
experience, a dose of laughing gas, or a series of responsive
readings from The Brothers Karamazov. The process of con-
version would be irrational were the monotheistic syndrome
acquired solely as an effect of a mystic experience or a read-
ing of Dostoevski, just as deconversion from the syndrome
would be irrational if, one fine morning, one simply woke up
without it. A rational deconversion occurs in consequence of
careful reflection upon the practical reasons that are found
inescapably to militate against it.
In this connection there can be no thought of necessary

or sufficient reasons, but only of good reasons which reli-
gious persons find themselves bound to take into account.

What is to count as a good reason is not something which
could be codified for us by theologians. In common practice

it is something which is recognized by ordinary reflective
persons, within a common
tradition, to be relevant. Apart
from a common, traditional form of fife which is accepted as
normative by most persons within a culture, the notion of a
good reason would have no meaning. But apart from such a
tradition it is doubtful whether the notion of reason in any
sense would have any significance.
Taken as a whole, the monotheistic syndrome cannot be
understood as a speculative hypothesis; it is not a "theory"
in the scientific sense or even in the metaphysical sense.
Such possibilities are precluded by the presence within it of
the moral claim which asserts that the almighty and omnis-
cient being is a perfectly good person. But they are also pre-
cluded by the religious claim itself. In avowing the theo-
logical thesis, as I should like to put it, we are not so much
asserting something about the nature of things as expressing
certain regulative attitudes toward the nature of things.
More briefly, we are testifying to a way of life or to the
foundational commitment of a way of life. Similarly, to reject
the theological thesis is not simply to deny a metaphysical
hypothesis about the origins of things but, more saliently, to
disavow the way of life which the monotheistic syndrome
prescribes. The theological thesis is thus a practical proposi-
tion, and commitment to it is a practical commitment whose
consequences for the person who accepts or rejects it are in-
calculably great.
THE SIGNIFICANCE OF PROPOSED SOLUTIONS

TO THE PROBLEM OF EVIL
We are now in a position to consider the significance of
solutions to the problem of evil. In turning to them, we must
re-emphasize here that what we have todo with is a prac-
tical, or existential, problem of faith and morals which is
only secondarily concerned with matters of fact. The incon-
sistency between the theological thesisand the ethical thesis
represents a conflict of attitudes toward reality and the con-
duct of life rather than a contradiction in our beliefs about
the nature of what exists. Hence the question of removing
such an inconsistency concerns essentially the possibility of
a consistent system of religious and moral sentiments rather
than the possibility of a consistent theory about the nature
of things.
In principle, there are many ways of seeking relief from
the problem of evil. Some of them involve primarily some
modification of the theological thesis. Others involve a modi-
fication of the ethical thesis. Whichever way is taken, the
inevitable result will be a modification of the way of life of
the person involved. Let us first consider those forms of re-
lief which do not entail a direct rejection of the ethical the-
1
God and Evil 19
sis. Some of them involve a radical deconversion from the

monotheistic syndrome and hence the adoption of a very
different attitude in regard to the object of religious faith;
others involve changes of a more limited sort.
The least radical way of relieving the problem of evil is to

find, or make, some excuse for the existence of evil which
modifies God's responsibility for it. But if this way is prefer-
able, since seems to leave the monotheistic syndrome

it
fairly intact, it is also much more difficult to render convinc-

ing precisely because of the claims regarding the nature of
God. It should also be borne in mind that it is one thing to
propose an excuse and another to accept it. No such proposal
functions as an excuse unless someone is prepared to accept
it. A proposed excuse is no excuse at all, any more than a
mere hypothesis is a statement of fact. Moreover, finding

an excuse is a very different sort of thing from finding a ten-
dollar bill or a cure for cancer. It does not involve looking in
a certain place, nor is it like making a lucky guess concerning
causal correlations which may happen to be confirmed by
observation. Rather, it is to give a counter practical reason
which defeats or extenuates or removes a charge of respon-
sibility. To excuse perform a distinctive act which is cor-
is to
related with acts of moral praise and blame. Where there is
no question of praise or blame, there can be no question of
making or finding excuses. And where there is a question of
excuses, there is also the question of praise and blame. A
valid or acceptable excuse serves to impugn the legitimacy
of blame or to sustain the legitimacy of praise. Invalid ex-
cuses formally leave open the question whether the praise or
blame is justified. In most situations, however, there are not
many plausible excuses to be made, and it frequently hap-
pens that there is either one excuse or none at all.
There are, I think, four main ways of making excuses for
a person: (a) by showing that the act in question was un-
avoidable; (b) by showing that, although the act was avoid-
able, it is justified in view of its consequences; (c) by shift-
ing responsibility, in whole or in part, to someone else; and
(d) by making an exception of the person in question.
Now, given the assumptions of the theological thesis, it is
hard to see how a plausible excuse of the first sort could be

made for God. Apparently he cannot be excused on grounds
of ignorance or inability. If such an excuse is made either
through some qualification of the concept of omniscience or
of the concept of might, then the conception of the holy or
worshipful itself already begins to undergo a sea-change with
consequences that may well be devastating to the delicately
balanced conception of man's religious and moral relation-
ship to God. For example, it might be argued, in the manner
of Leibniz, that any universe created by God must be the best
of all possible worlds and that for whatever evils that exist
there must be, a priori, a sufficient extenuating reason. Logi-
cally, and perhaps metaphysically, God could have created
some other state of affairs, but since he is a perfect being, it
is theologically impossible that he should create any other
universe than the one which exists. In short, if God exists, his
act in creating this universe is unavoidable. The effect of
such an excuse, however, is simply to reaffirm the absolute
—
benevolence of God which is the very point at issue. It
clarifies the aspiration of the monotheist, but
it does not nec-
essarily remove moral qualms that the world is ethically

his
out of joint. If, like Pangloss, he does manage to accept the
excuse, he must now view the evils of the world in such a
fundamentally different light that his whole conscientious at-
titude toward them is virtually transformed. That is to say,
if he is consistent, he is bound to accept in advance whatever
happens to exist as something which is at once unavoidable

and good. I shall not here remark upon the merits of the
quietism and optimism implicit in such a view. But it is
doubtful whether anyone deeply involved in the problem of
evil will find it easily acceptable.
Initially, the second sort of excuse may seem more promis-
ing, particularly if we restrict ourselves to the question of
moral evil. Thus, it might be argued that, although God may
have a certain collateral responsibility for moral evil, since
he granted freedom to man knowing in advance what use
man would probably make of it, moral freedom is neverthe-
God and Evil 193
less a good so great that

outweighs any possible evils that
it
may attend it. is saved by claiming

In short, God's goodness
that God's act, in creating man as morally free, has good
consequences that outweigh all admittedly evil consequences,
and that although he shares, in one sense, responsibility for
moral evil, his goodness as a person is not thereby to be im-
pugned. This excuse is of interest because it commits him
who offers it not only to the intrinsic worth of moral freedom
but also, comparatively speaking, to the insignificance of
happiness, aesthetic values, and, indeed, all the goods that
are usually classified under the heading of "welfare." But
again, it is a question whether men of ordinary good con-
science will be able to make this commitment. If they do,
another profound shift will doubtless occur in their scheme of
values. It should be noted, however, that such an excuse does
not touch the question of natural evil, including the in-
voluntary ills that innocent persons are made to suffer in
consequence of the evil acts of others. Here, one can imag-
ine, those afflicted with a sense of estrangement or forsake-
ness ( My God, my God, why hast thou forsaken me? ) or out-
raged at the "inexcusable" sufferings of their loved ones are
not likely to be edified by any attempt to show the greater
good inherent in the consequences of these evils.
Doubtless there are some who would hold such a justifi-
cation of moral evil to be defective precisely because, al-
though it removes God from blame, it does not entirely re-
move him from responsibility. From this point of view, so
long as God is held to be responsible in any way at all for
moral evil, our faith in him is imperfect and our own sense of
freedom and moral responsibility is incomplete and im-
mature. Thus it may be argued that we ourselves must share
the total burden of responsibility and guilt for our acts, and
that God's act, in making us free, has nothing to do with our
evil choices.
What is the spiritual significance of such a view? Now, in
ordinary circumstances, when a person is relieved entirely
of responsibility for a particular act, his ethical connection
with it ceases, and any question of praise or blame of him
so far disappears. Ethically, so to say, he becomes irrelevant

to it. If, then, we alone are to be held responsible for our
actions, if our responsibility cannot be shared, then not only
any blame but also any praise is ours alone. How could we
consistently praise God from moral evil, if we
for saving us
cannot share with him the blame when we do evil things? If
he has no hand in the evil, how then could he be said
consistently to have a hand in the good? When we commit
murder, the sin is ours. But then, when we are loving, kind,
or just, must not the credit and the praise be ours as well?
In that case, however, any reciprocal ethical relationship be-
tween ourselves and God would tend to disappear. Corre-
spondingly, it becomes increasingly hard to see how he
could still be addressed as a "Thou" to whom we are in any
way morally beholden.
The most radical way of excusing God from responsibility
for evil is simply to make an exception of him. This sort of
thing is done in ordinary life when someone is pardoned for
a wrong act which he has committed. A man is also some-
times excused, whether properly or not, for acts involving a
moral fault if, for example, he is acting in some publicly ac-
knowledged capacity which carries with it non-moral re-
sponsibilities that, in the circumstances, arethought to be
overriding. In the latter case, the excuse simply places the
individual, for the nonce, beyond moral good and evil.
Were such an excuse made, for whatever reason, in the
case of God, it would undoubtedly involve a modification of
the monotheistic syndrome and with it a modification of the
sense in which monotheism could be considered an
still
ethical religion. As God, perhaps it

for pardoning or forgiving
is possible that we might try to do this, but not without im-
pairing God's value to us as an ethically exemplary being,

and not, therefore, without diminishing our sense of re-
sponsibility to him. Normally, it has been supposed that the
relation of forgiveness should go the other way around: we
are to forgive one another, even as God forgives us.
In general, any form of theodicy, the aim of which is to
justify God's ways to man, is bound to appear implausible to
God and Evil 195
anyone who has an overwhelming sense of the natural evils

that abound in the world; and at best, it only serves to ex-
tenuate God's responsibility for moral evil. Even so, the con-
—
sequence of having to make excuses for God to "save his
face," as it were— is to jeopardize our own sense of obligation
to him and to render his goodness inoperative as a moral

ideal. Religiously, the almighty being might still be con-
ceived as God, but he would no longer be looked to as the
source of moral goodness and as the being to whom, in the
moral sense, "all praise is due." Briefly, theodicy nearly al-
ways increases the sense of the distance between man and
God, and the remoteness of the moral life from the life of
religious worship. In some people it may also reinforce a
sense of complacency, resignation, inevitability; but they are
precisely those who are least likely to be affected by the prob-
lem of evil. Conversely, those who are affected by the problem
are most likely to be repelled by theodicial excuses. In so far
as they do accept them, their excuses for God serve to re-
inforce the sense of personal guilt which may, in the end,
become so overwhelming that nothing is left of their sense of
—
personal dignity save the bare freedom to go on sinning.
Several other ways of relieving God of responsibility for
evil remain, but they are all more drastic, since they involve
a more radical modification in the conception of God's nature
and hence in the attitudes that are possible toward him. All
such ways of overcoming the problem of evil involve a funda-
mental breakdown of the monotheistic syndrome or, what
comes to the same thing, a fundamental attenuation of the
traditional monotheistic faith. Nor can they be effected with-
out in some measure incurring a liability to deconversion
from monetheism.
The most radical, but in some ways easiest, way out is to
renounce one or more of the attributes ascribed to God
by the metaphysical claim. One may cease to hold that God
is almighty or that he is omniscient, or may cease to regard
omnipotence or omniscience as necessary attributes of holi-

which is worthy of piety or worship. In so doing,
ness, of that
the problem of evil is automatically relieved, since a God
who is not almighty or who is not omniscient need incur no
responsibility for evils that he may not have anticipated or

could do nothing to prevent. Giving up the metaphysical
claim, however, does not require renunciation of faith in
God's goodness; nor does it preclude us from still worship-
ping him as the holy of holies. But it does affect our attitudes
and acts of faith in other important ways. For while God
may be worshipped, claims can no longer be intelligibly
still
addressed to him and, at least in one sense, praying to him

becomes correspondingly poindess. Likewise, it becomes in-
creasingly difficult to view the whole of creation in a reli-
gious spirit as the providential work of deity. But, most
seriously, the sense of absolute dependence upon God tends
to disappear and, as one realizes that the amelioration of
one's lot depends, not upon Grace alone, but upon one's
own efforts, God ceases to appear in the guise of an all-
powerful father, and becomes rather a kind of outsized
brother who has his own troubles with evil, even as you and I.
I, for one, respect those contemporary theologians who
candidly accept the notion of a finite God. If such finitistic

theologies are still involved in grave metaphysical diffi-
culties, they at least enable the believer to preserve his moral

integrity and his religious devotion to a God who can be re-
garded with a straight face as a good person. They are pos-
sible, however, only to men of great emotional maturity and
stability for whom, so to say, the trauma of birth has in some

measure been overcome. They are not appealing to those for
whom religious piety is possible only to the degree that the
needs of emotional dependency are satisfied.
Another way of relieving the problem of evil is to qualify
or else to give up altogether the moral claim concerning the
goodness of God. To some, for whom religion means ethical
religion, such a possibility may appear repugnant; but log-
ically there is nothing to oppose it. There is no logical con-
nection between the metaphysical and the moral claims of
the theological thesis, and it is entirely possible to regard an
omniscient and almighty being as God without investing him
with the attributes of a perfectly good person. But if one
God and Evil 197
does accept this alternative, God automatically loses his

moral significance, and the autonomy of the moral life must
now become a reality. Thus, while God's "commandments"
may still be obeyed, one's obedience to them is no longer a
consequence of one's faith, or, to say the same thing in other
words, they are obeyed not because God commands them
but because they are deemed right and good on their own
account, regardless of the fact that he commands them.
From this standpoint, the religious life now detaches itself
altogether from the ethical with consequences which

life,
are incalculable to both. The appeal to God's will can no

longer be used in justification of a moral principle or in ex-
tenuation of a practice which, on other grounds, might seem
wicked or nefarious. It must be confessed, moreover, that
the deification of the metaphysical attributes as such is likely
to appear repugnant to an ordinary moral consciousness,
and that once God's goodness as a person has been effec-
tively challenged, the whole interest in his metaphysical at-
tributes is likely to collapse.
The most drastic way out of the problem of evil is to give
—
up altogether the belief in God the belief, that is, that
there is anything worthy of worship. But this way of putting
the matter is ambiguous. For one may renounce the belief
that any actual thing is worthy of worship, without ceasing

to have attitudes which may be deemed religious. Thus, for
example, philosophers like Comte or Dewey appear to have
thought that while no actual thing is worthy of worship the
moral ideal is itself worshipful. This is not the place to enter
into detailed discussion of the concepts of existence, ac-
tuality,and being. There is, however, a quite proper sense
in which one might be able to say that there is something
worthy of worship, or even that something is God. without
at the same time implying by this that there is a thing or a
person that is God. The term "something" does not mean
"some thing," and the contention that something is worthy of
worship or that something is God does not entail that there is
a thing, a substance, an actual entity which is worshipful or
God. It is owing entirely to the exigencies of our own
Western religious tradition, and to the monotheistic syn-

drome which goes with it, that we would consider it mis-
leading to claim that there is a God, an object worthy of
worship, while at the same time denying that there is any
thing or any person that is worshipful and toward which our
piety is due. And it is merely because of this that it has not
been customary for naturalists, materialists, or positivists to
say that they believe in God or that God exists. This is per-
haps nothing more than a sign of the parochial limitations
of Western theology which, all too frequently, mistakes the
meaning of the so-called "atheistic" religions of the East.
From a logical point of view, there is nothing whatever to
preclude the possibility of a person sincerely affirming his be-
lief in God while at the same time denying that God is a
person, a substance, or a thing.

This possibility suggests a way of modifying the mono-
theistic syndrome itself in such a way as to solve, once for
all, the problem of evil. Thus, one might adopt the view that
God, as such, is completely unknowable, and that as such no

positive attributes may validly be imputed to him. From
this standpoint, God's nature becomes a complete mystery,
and the worship of "him" as an "other," not ourselves, ceases
altogether to be the worship of an almighty or good person
who creates the universe out of his infinite goodness and
who gives to man the commandments by which he is to live.
This is the move which has been taken, now and then, by
"pure" mystics and by men of "pure" faith. But it is a form
of adoration or worship which is virtually devoid of content
and which, at last, merely battens upon its own subjective
intensity. Saying nothing, it commits us to nothing; affirming
nothing about God, it divests God of any significance for the
conduct of life. It solves the problem of evil, but only
by removing the religious life altogether from contact with,
and hence possible contamination by, the moral life. Pre-
cisely because it views God as utterly transcendent or, as
some theologians put it, as sheer "transcendence" itself, it
loses all relevance to the ordinary problems of human action.

Such, I take it, is at bottom the theological position of
God and Evil 199
Kierkegaard. But the spiritual remedies which it proposes

are desperate remedies which ordinary persons are not likely
to be able to accept. What remains of monotheism when it is
accepted is anybody's guess —indeed, it is not for nothing that

Kierkegaard himself has been sometimes called an "atheist."
No doubt some precedent for the view can be found in the
biblical literature. But surely the drift of what most people
think of as biblical monotheism is incompatible with it, and
surely it is not the view ofGod which is normally entertained
in our churches and our synagogues.
FURTHER SOLUTIONS
ways of relieving the problem of evil
All of the foregoing
involvesome tampering with the theological thesis and
hence some degree of deconversion from the monotheistic
syndrome. Traditionally, however, relief has frequently
been sought through a modification of the attitudes ex-
pressed by the ethical thesis that something in the created
universe is evil. It is this way out which most obviously
shows how morals and religion may come into conflict and
how, in turn, moral sentiments may be modified to meet the
exigencies of religious faith.
The most drastic, if also the most implausible, way of deal-
ing with the problem of evil from this standpoint is simply
to deny the ethical thesis altogether. On this view, which
most men would regard as highly immoral, there simply is
no evil, and when someone says that something is evil he
speaks falsely: evil, that is to say, is merely apparent, and the
common view that such things as earthquakes, insanity, and
cancer are evils is illusory. This, we have been told, is the
position taken by Christian Scientists, but I do not vouch
for it. In any case, it is hard for an outsider to see how such
a point of view could be consistently maintained in prac-
tice. No doubt a sufficiently resolute optimist might say, for
the nonce, that God's in his heaven and all's really well with
the world. But it is not something which can seriously be as-
serted by anyone who takes a dim view of pain, loneliness,
zoo Reason and Conduct
and cruelty. No ordinary Jew or Christian, so far as I can see,

could entertain it for more than a moment, for it makes
nonsense of the moral life. If nothing is evil, choice is point-
less, and responsibility has no meaning.
A subtler view involves a persuasive redefinition of evil
which conceives it as merely a privation of being. According
to this view, whatever is created by God is good and hence
worthy of praise. Evil, no doubt, is a necessary consequence
of finite or created being, but only of its finitude, not of its
being. Thus, so far as they are, pain, suffering, and guilt are
good. The evil in them is only that of a limitation or privation.
At this point, it is easy to become entangled in a web of
words. We are interested here only in the moral significance
of such a view, that is to say, in its bearing upon our spiritual
attitudes toward the world and toward the conduct of life.
Formally, everything seems to remain just as it was be-
fore, since, in one sense, it is still possible to say that some-
thing is evil. But materially, a complete sea-change comes
over the moral life. Thus, for example, so far as pain or suffer-
ing are positive facts of experience, they are to be approved
of as good, as something, that is to say, which ought to ex-
ist and which should be affectionately endured or, more
strictly, rejoiced in. To that extent, the will to remove suffer-
ing and to root out pain, as well as the conviction that,

simply per se, pain ought not to exist, must be regarded as
forms of impiety. The man who worships God as good and
accepts His creation as part of His divine providence, must
so far acclaim pain and suffering as intrinsically good. It is
only in so far as pain indicates some privation of being,
some privation of a higher perfection which is potentially
ours, that it can be regarded, albeit misleadingly, as evil.
Strictly, it is merely an indication or sign of evil; what is
evil is simply the accompanying privation.
To many moralists the quietism implicit in this view has
seemed obvious. It is not my purpose here to evaluate it
either from a moral or from a religious viewpoint. It is un-
doubtedly the view of many so-called saints, and those who
are taken with saintliness will undoubtedly be attracted to it.
God and Evil 201
My point is that it is not an ethically neutral position, and

that it does not leave our ordinary moral attitudes just where
it found them. Moreover, strictly speaking, it cannot be
regarded as a final solution of the problem of evil. For it still
countenances the assertion that something, namely, privation
of being, is evil. Indeed, from the point of view of those who
are faced with the problem of evil, the privation theory
merely forces the issue onto another plane. The problem,
now, is to justify creation itself, which reopens the question
of theodicy in its most radical form. It may be argued that it
was always possible for God to choose not to create anything
at all, and that if he could create something only by bring-
— —
ing evil or privation into the world he should not have
created anything. Here we find the problem of evil in per-
haps its most poignant and intractable form. For even if the
individual can make his peace with pain and the sense of
moral guilt, he may still find existence itself stale, flat, and
unprofitable, and he may still be torn between his desire to
loveGod and his unqualified loathing of the world which
God has created.
Here we face the very issues implicit in Professor Tillich's
discussion of what hecourage to be." It is essen-
calls "the
tial to bear in mind, however, that the courage to be is not
something which can be required by theological pro-

nouncements. For those who lack this sort of courage, the
problem of evil is not remotely relieved by pointing out
that evil is a necessary consequence of finitude, and hence of
all created being, for it is the very evil of creation itself
which is now in question. Why should anything exist? Such
a limiting question is not meaningless, but it cannot be
answered by the ploys of theodicy. It can be answered, if at

all, only by a change of heart. But who shall say to an-
other that such a change of heart is good? Even so, recovery

of the courage to be does not necessarily involve a recon-
ciliation of faith and morals, since it is entirely possible that,
in recovering, a man might still, in his heart, condemn as
evil many positive features of created being, with the conse-
quence that his faith in the goodness of God continues to
conflictwith his conviction that at least some of God's works

ought not to be at all.
THE PROBLEM OF EVIL AND THE LIMITS OF REASON

Up to this point, I have proceeded on the tacit assumption
that the problem of evil is itself an unmitigated evil, that it
is something which, at least from a logical point of view, is

intolerable. I have assumed, that is to say, that the accept-
ance of an inconsistency, and especially a practical incon-
sistency, is wrong and that it ought to be overcome. In effect,
this means that, like most moral philosophers, I have taken it
for granted that self-consistency and integrity in the spiritual
domain is the consummation most devoutly to be wished. But
is it? Why should the demand for consistency be regarded as
categorical or as taking precedence over every other re-

quirement of the moral or religious life? Within the moral
sphere itself, we are frequently faced with conflicts of prin-
ciple which can be removed only by subordinating one prin-
ciple to another and which, therefore, can be removed only
by disavowing some hitherto fundamental moral commit-
ment. In such a case, is it reasonable to insist that an in-
dividual ought to modify his principles? Without begging
the question, what sort of reason could be given that would at
the same time satisfy the requirements of logic and those of
conscience?
Suppose, then, that someone found himself faced with the
problem of and that even after the most deliberate and
evil
painstaking reflection he still found no answer to it. Suppose
also that he found that he could bring himself to reject or

qualify neither the theological thesis nor the ethical thesis.
No doubt, such a person would find that his religious aspira-
tions and hismoral intuitions cannot be reconciled and hence
that at the very heart of whatdeepest and most sincere in
is
him an insuperable and unremit-

as a person, there remains
ting tragedy of spiritual self-division. Shall he be told to
fish or cut bait? But from what standpoint could such a re-
quest significantly be made? If we appeal to the principle
God and Evil 203
of consistency or integrity, we are evidently appealing to

something which, even if he is sensible of its proper claims,
requires him to renounce his profoundest loyalties to cease, —
that is, to be the very person he is. To reply to him, as Pro-
fessor C. I. Lewis might, that if he continues self-divided
he will be sorry, is not to tell him something he does not al-
ready know. Of course he will be sorry; he is already sorry.
But what, in the name of goodness and of God, is he to
do?
In saying this, I do not wish to be misunderstood. I have
been trying only to illustrate certain approaches to the prob-
lem of evil in order to see what their existential or practical
—
import really is not, on my own, to advocate one solution
above another. Study of the problem of evil is instructive in
many ways. It helps us to see, in the first place, how ques-
tions of logic may moral and religious life as
arise within the
well as within science or mathematics. The very possibility
of a problem of evil makes plain the fact that questions of
consistency are not impossible in the domains of religion
and morals. Such questions, I have argued, are, at bottom,
and the reasons that may be offered in attempting
practical,
to answer them must provide motives to influence the will,
or heart, or they are nothing. Any arguments in this sphere
have a point only in so far as they effectually dispose us to
modify our active religious or moral commitments. Any re-
ligious or moral dialectic must be, in Hegel's phrase, a
"logic of passion."
In the second place, the problem of evil makes us forcibly
aware that philosophical theories which insist upon the neces-
sary "autonomy" of moral and religious principles cannot be
sustained. This is not to say that it is logically impossible to
divorce morals from religion and to separate questions con-
cerning what is right or good from questions concerning
what is worshipful or holy. seems evident to the humanist
It
that questions of right and wrong are answerable without re-

course to the will of God. What is good is good, and the fact
that God wills it cannot make it more so. What he forgets
is that, for the theist, God's will alone certifies that anything
is good at all. Here, however, is no mere problem of an-

alysis which closer inquiry into the meanings of moral and
religious expressions might be thought to solve.
Too often such a problem is presented as an external con-
flict between two distinct ways of life. But for anyone in-
volved with the problem of evil it is an internal problem,

since in affirming the ethical thesis in the teeth of the theo-
logical thesis he is so far acting as a moral autonomist who
accepts as final the deliverances of his conscience, whereas
in affirming the theological thesis (which itself involves a
moral claim) he is accepting the will of God, whatever it is,
as that which alone makes good things good. Such a person
is caught within himself in a practical conflict of religion
and morals. But it would be more accurate to say that he

is caught in a conflict between ethical independence and
ethical dependence, between a conscience which presumes

to judge for itself what is good or evil and an ethical faith
which depends upon the will of God to authorize what is
to be believed in the moral sphere.
The study of the problem of evil is instructive, finally, in
that it enables us to see with great clarity what it can be to
question the existence of God and how, within the religious
context itself, such a question is bound to arise. I have never
doubted that the word "God" gets its primary meaning from
the part it plays in religious discourse. And I not only ac-
cept but insist upon the thesis that religious speech and lit-
erature provide the paradigmatic uses of the term which it
is the business of the philosopher to explain. But religious

discourse is not the discourse of angels who do not know the
meaning of doubt, but of men for whom doubt is a con-
dition of life. To ask whether there is a God constitutes the
fundamental trauma of the religious life itself, just as the
question whether anything is good is the great night-
mare of the moral life. In the case of our own traditional
forms of monotheism, this trauma is brought on directly by
the problem of evil. Pending resolution of the problem, or
our absolution from it, the possibility of answering "No" to
the question of God's existence hangs immediately in the
God and Evil 205
balance. Religious faith is not necessarily dependent upon

the confirming evidences of reason. But it may be impugned
by contrary evidences which reason recognizes to be in-
compatible with Although it is perhaps possible that
it.
men of faith believe or wish to believe that the theological

thesis is true, it is as certain as the proposition that two
plus two equals four that the thesis is falsifiable. What makes
it certain is Augustine's ancient question: "Whence, then, is
evil?"
PART TWO
[X]
Commonsense Ethics
and Ordinary Language
ORDINARY LANGUAGE AND LIMITING QUESTIONS

Future historians of ideas may come to regard the first
well
of the twentieth century as "The Age of Symbol." As applied
to the fine arts, to literature, to psychology and psychiatry,
and to social studies in general, this epithet connotes a re-
fusal to accept appearances at face value, an insistence upon
interpretation, indirection, complexity, and depth. As ap-
plied to philosophy, however, it will probably suggest to
some minds that ours is "The Age of Superficiality" par ex-
cellence, a period that has witnessed the possibly final aban-
donment of philosophy's ancient quest for understanding and
assurance concerning the nature of things. But to those who
begin to discern the ulterior ramifications of the study of
signs, I. A. Richards will scarcely seem to be exaggerating
when he it is "the most fundamental and extensive
says that
and not a mere "preliminary or preparation
of all inquiries,"
for other profounder studies." For them, philosophy, con-
ceived as the analysis of symbols, is for the first time coming
into its own. No longer required to provide the poor man's
science or the intellectual's substitute for religion, it is finally
free to fulfill the tasks of logical analysis which are proper
to it.
There are however, various ways of approaching the anal-

ysis of signs. And at the present time there are several schools
of "philosophical analysis," each as bitterly opposed to the

others' conceptions of its aims as all are jointly disapproving
of traditional metaphysics. One influential group, deriving
primarily from the earlier works of Bertrand Russell and Lud-
wig Wittgenstein, includes most of the logical positivists. This
group is mainly interested in mathematical logic, in the meth-
odology of exact science, and in the formulation of codified
"ideal languages" from which all vagueness, ambiguity, and
reference to unobserved entities have been eliminated. For
them the absence of clearly stipulated definitions and the
presence not only of multiple meanings but, worse, of mul-
tiple modes of meaning, make natural languages inadequate
vehicles for the precise communication of logically coherent
descriptions of fact. On the whole this school has been in-
different to the noncognitive, non-scientific uses of language.
And it has tended, somewhat indiscriminately, to consign
moral, poetic, religious, and "philosophical" discourse to the
semantic dustbin of "expression" or "emotive meaning."
Meanwhile, particularly at Cambridge and Oxford, an-
other school has arisen to challenge the methods of the posi-
tivists.Sometimes referred to as the "philosophy of ordi-
nary language," its somewhat discredited godfather is G. E.
Moore, whose powerful dialectic and passionate devotion to
"common sense" provided the initial incitement to the gen-
eral revolt which has taken place in England against the ob-
scurity, the jargon, and the artificial technicalities of the
classical metaphysical systems. Moore not only stimulated a
renewed interest in and respect for ordinary language, but
also showed by his own example the possibility of conducting
subtle and exact analyses without ever resorting to a logical
apparatus beyond that afforded by natural language itself.
More recently, it has been the stimulating informal discus-
sions of the later Wittgenstein and the writings of such fol-
lowers as John Wisdom and Gilbert Ryle that have steered
the philosophy of ordinary language into its present course.
Moore was not hostile to metaphysics as such, and in his
own ethics he was quite capable of asserting the existence of
unanalyzable "non-natural" qualities and intuitively neces-
Commonsense Ethics and Ordinary Language 211
sary moral truths. But Wittgenstein and his friends are as

suspicious of metaphysical "double-talk" as are the positiv-
ists, although for somewhat different reasons. Thus, for ex-
ample, in his brilliant The Concept of Mind, 1 Ryle patiently
attempts to uncover the categorial confusions which, as he
believes, lie at the heart of such hoary philosophical issues as
"the mind-body problem." Moore had often been content
simply to bellow, "Of course I know I had my breakfast this
morning," or "Certainly, this is a hand; who could doubt it?"
His younger contemporaries are more interested in revealing
the sources of the linguistic confusion that give rise to such
"queer" philosophical doubts than in asserting the obvious
commonsensical truths which should dispose of them.
On one point the later Wittgenstein and his group vio-
lently part company with the other school of analysts. They
respect and defend ordinary language, but, as Max Black has
recently pointed out, they also regard those who are pre-
occupied with the construction of "ideal languages" as guilty
of the fundamental philosophical blunder of "treating lan-
guage like a calculus." Aware of the great richness and com-
plexity of ordinary language, and perhaps because of their
more adequate training in the humanities, they realize that
modes of meaning other than the logical or scientific may
have their own implicit standards of precision and relevance.
And what is more important, they have tended to supplant
Moore's question "What does it mean?" with such questions
as "What is its function or role?" and "What are the distinc-
tive criteria for judging the success or failure of symbols in
the very different roles which they are called upon to play?"
Such an approach to the study of symbolic forms is, in my
judgment, a stimulating and fruitful way of considering
them. But it has its own limitations, and in the hands of its
more pedestrian advocates, it can be exceedingly stuffy and
often vague. Especially when, somewhat in the
is this so
manner of Dr. Johnson, all violations of ordinary usage are
condemned out of hand on the sole ground that ordinary
usage is the correct usage. There is no a priori reason why
1 G. Ryle, The Concept of Mind (London: Hutchinson, 1949).
ordinary language should invariably be capable of adequately

expressing what we may wish to say when we pass beyond
the middle-sized problems with which it is designed to deal.
Be this as it may, the philosophy of ordinary language has
shown how easily and, in many instances, how unnecessarily
we can pass into paradox and pseudo-problems when we mix
up our categories or misconceive the different roles of words
in common speech. Mr. Stephen Toulmin's recent An Exami-
nation of the Place of Reason in Ethics2 is an impressive ex-
ample in the sphere of moral philosophy of how, by shifting
the question, the philosopher of ordinary language may help
to remove the "mental cramps" that for a generation have
prevented subtle and careful thinkers from resolving the is-
sues which divide them.
Concentrating exclusively on the "meaning" of such terms
as "ought" and "good," most moral philosophers continue to
debate interminably the question of whether they refer to
objective qualities or are subjective expressions of approval
or disapproval. Mr. Toulmin asks, in effect, that this question
be deferred, and that we consider instead the kinds of "rea-
sons" that are ordinarily regarded as relevant or acceptable
in arguing amoral question. In the end, he maintains, we
may perhaps better understand the meanings of moral judg-
ments by following this course than by attacking this latter
problem head on, without regard to the special role played
by such judgments in the larger system of activities of which
they are a part.
In the first part of his book, Toulmin explains what he be-
lieves to be the each of the three main
fatal flaws implicit in
traditional approaches to ethics. The first, which he calls "the
objective approach," is vitiated from the outset by an uncriti-
( [y cal assumption that when we assert that something is "right"
we are simply ascribing to it an objective property in the
same way as when we assert that it is "yellow" or "cold." But
in fact ethical terms just do not function in this way. And so
the objectivist is forced to assume an entire battery of special
2 Stephen Toulmin, An Examination of the Place of Reason in Ethics
(Cambridge, Eng.: Cambridge University Press, 1953).
categories such as "non-natural" qualities and a priori syn-

thetic propositions, and special "faculties" of "moral intuition"
and "practical reason," in order to account, even in principle,
for the distinctive features of moral discourse. Even so, his
whole elaborate construction in the end falls to the ground

when we ask the simple question, "What is the distinctive
quality 'goodness' supposed to be common to hu-
which is
mility, pleasure, friendship, and promise keeping?" For when

we look, even with the "eye" of the mind, this quality always
eludes us.
The basic error in this approach,Toulmin believes, is the
logical blunder of assuming that when one person asks an-
other whether a given course of action is right, he is inquir-\
ing about a property, when in fact what he really wants is a }
reason_ for deciding what to do. In m oral discourse we do not /
expect merely to be believed when we inform someone that
an act he is about to perform is wrong; we expect also, if our
statement goes unchallenged, that we will be construed as
providing a valid reason for not performing the act. And if
someone tells us that he doesn't know what Tightness is, we
don't accuse him of not knowing the meaning of a word;

rather do we expect him to break his promises, or to lie, or
steal.Toulmin points out that in so doing we implicitly recog-
nize that what is meant by "knowing what Tightness is" is
very different from "knowing what redness and hardness
are.
Disillusioned with an approach that invariably lands us in
Queer Street, many philosophers have taken the ^subjective (Qj)
approach." In this view, the sharp distinction, between
values and such subjective relations as "desired" or "ap-
proved," which is insisted upon by the subjectivists, is
denied. Despite obvious weaknesses, which in their turn the

objectivistshave not failed to indicate, the subjective ap-
proach has points. For one thing, it stresses the obviously
its
close connection between "desirable" and "desire," and be-

tween "value" and "satisfaction." For another, it also appears
to provide an intelligible account of the genuine variations in
the ethical judgments and standards of persons who are in
other respects equally well-informed. Nevertheless, it has its
own fatal flaw, for it is incapable-of-accounting for_rahat we

actually mean by a "good reaso n" for desiring or approving
something as distinct from a "reason" which merely causes us
to desire or approve In a word, the subjectivist inevitably
it.
K]nr^^ji«; tl^pi-.iQrir; hntwnnn nbligntirm and inclilia£ion_and

between what Hu Hif^rm rall^H "j"<*ifying" and HfijscJ&Dg"
reaso ns.
We are left, therefore, with our central question still un-
answered: "What makes an evaluative inference valid or in-
valid, relevant or irrelevant?" Both the objectivist and the
subjectivist fail to provide an answer because they both as-
sume that ethical judgments can only contradict each other if
they refer to an objective property of the object in question,
y ) and that unless they do refer to such a property, they must
'express some psychological state of the speaker, in which
case no significant contradiction can possibly arise. But why
should we assume that these are the only alternatives? It is
just this assumption which is challenged by the ".imperative
c* appro ach." In this view ethical judgments are essentially dis-

guised commands, and they oppose one another simply as
conflicting prods to action. Like the other approaches, how-
ever, the imperativist accepts without question the thesis
that there can be no question of validity which is not strictly
factual or, in the narrow sense, "logical." Accordingly he
regards the whole question of "justification" in ethics as
senseless. Commands cannot be logically derived from state-
ments of fact; nor can they be intelligibly said to be verifia-
ble as true or false. Hence the only "reasons" which can be
given for obeying them are purely "exciting reasons."
The flaw in this approach is that it fails utterly to compre-
hend that, whereas in the case of ordinary commands argu-
ments are logically irrelevant, this is not so in the case of
y\ particular moral judgments. In the latter case, it is always
( "reasonable" to ask why we ought to do what they prescribe.
/And what we want is not just any answer that will "do the
trick" of causing us to agree with the speaker, but a real rea-
son for agreeing even when we do not agree.
Toulmin himself is a pluralist in the sense that he believes

( as I think, correctly ) that the sense of ethical terms tends to
shift with its context. There are occasions when "Good!" may
mean little more than "Hurrah!" But, especially when we are
pressed for an explanation or justification, the sense of ethical
terms gradually shifts to something which is more "objective"
or impersonal.
The conclusion which Toulmin reaches in the course of an
ingenious and frequently convincing analysis is that while
there are important analogies between the functions of rea-
son in science and in ethics, there are also some significant
differences, which indicate essential differences in the func-
tions of the two types of discourse. Both are concerned, in
some sense, to distinguish between appearance and reality;
but what counts as "real" is very different in the two spheres.
In ethics, the purpose of judgment is not to correct our mis-
taken identifications and expectations but to alter our senti-
ments and to correct our behavior. So far Toulmin roughly
agrees with the imperativists. But he also recognizes, as they
latter do not, that something of which we express our ap-
proval may not be "really good."
There are two ways in ethics, according to Toulmin, of
showing another person why, as we say, we really should not
perform a certain act: (a) we may explain that the act con-
travenes some part of the customary moral code of the com-
munity to which he belongs; or if this fails, we may (b) fur-
ther explain that the consequences of the act would be likely
to cause other members of the community some serious in-
convenience or suffering. These types of "reason," says Toul-
min, simply "cry out" to be called "moral." And surely he is
right. Notice, however, that he is not saying that "good" is
descriptively synonymous with "approved by the commu-
nity" or "conducive to happiness." Rather he is asserting that
when we attempt to provide a relevant reason for a particu- / >
larmoral judgment only two main types of supporting argu-

ment are regarded as ethically valid.
In this view, an immoral person is primarily and literally
an outlaw. To advocate, as Nietzsche prescribed, living be-
yond "good" and "evil" is simply to declare oneself in effect

impervious to the kinds of reason which morality acknowl-
edges. In principle, one might be none the less kind or be-
nevolent for all that. But Toulmin also recognizes that, at
least in "open" societies, there is a legitimate ethical proce-
dure for criticizing the moral code of the community itself.
And he sees that breaking the moral code must be sharply
distinguished from disinterested moral disapproval of it. It is
important to recognize, however, that when one questions
whether an act which is in fact prescribed by the prevailing
moral code is really right, one is going outside the normal
ethical universe of discourse of one's community. In such a
case one is challenging not the propriety of some particular
act, but the whole moral code itself.
Even though this sort of challenge is morally permissible,
at any rate in open societies, there are clearly limits to the
questions which may be validly asked about the right and
the good. Thus someone should ask as a matter of princi-
if
ple, "Why should I be moral?" there just is no valid answer to

„the question. As Toulmin says, there is simply no room
J within ethics for it. It resembles nothing so much as the com-
|
pulsive "Why?" of a child, who will accept no valid explana-
I
tion that we can give. For this reason, Toulmin thinks that
"philosophical" attempts to justify morality as a whole by
referring it to some ulterior metaphysical "ground" are bound
to be pseudo-explanations. Like attempts to justify our belief
in the external world or in the existence of other selves, they
invariably pass beyond the bounds of sense, precisely—be-
cause we have.no notio n of the kind of reason which_would
co nstitute a valid an swer.
And yet Toulmin, in the manner typical of the younger
members of his school, does acknowledge the importance of
what he calls "limiting questions," irritating and misleading
as they often are. Psychologically they help us to adjust to
the world, just as the explanations of science help us to
understand it. He recognizes, in short, that there are pro-
foundly important human problems which pass understand-
ing.
ON SAYING WHAT CANT BE SAID IN ETHICS
Just at this point, however, certain general doubts about the

adequacy of Toulmin's analysis begin to obtrude. If he is
right, then limiting questions about the merits of morality as

a whole or about the correctness of moral reasoning as a
whole necessarily involve an abuse of the ordinary language
of conduct. Hence, while he grants them a certain psycho-
logical significance, he is obliged also to regard both the
questions and the answers we may make to them as lacking
in any literal, rational sense. That is to say, they make sense,
according to him, only in the way a tic or a fetish makes
sense to a scientific observer who understands it as a symp-
tom of some psychological ailment. But they make no sense
as ethical questions and answers, since, on Toulmin's ac-
count, they seek to say what can't be said. As it happens,
they involve the use of words, but they do not employ the
words in question as words, just as eating peas with a knife
involves the use of a knife, but not the use of it as a knife.
I amconvinced that such a view of the matter cannot be
sustained, either as a theory of moral discourse or as a partial
theory of the common language of which moral expressions
are a part. Let us first consider the latter point. Now we may
agree_±hat^when philosophers become involved in general^
category mistakes, as they do when they treat ethical predi-
ca"tres-~n3£g "good" as names ^LlpfiiXieptuaLûalities like
"warm' or "yellow," they systematically misconceive and mis-

represent the linguistic function of moral judgments. But it is
not at all clear that such mistakes are involved when we ask
limiting questions. In such cases, it seems rather that we are
merely extending the application of such terms as "good"
and "right" beyond certain contexts in which they are famil-
iarly employed. What is involved here is not so much the
confusion of expressions of one logical type with those of an-
other, but rather the application of terms of a certain type in
circumstances for which there are, or appear to be, no well- \
defined precedents. Now it may be granted that in stretching'
the application of a term in this way we may temporarily

lose our logical bearings. And when this fact is pointed out we
may find that we no longer are so ready to press our ques-
tions to the limit. But this is not inevitable. Toulmin's theory
of language evidently assumes that the rules governing the
use of ordinary expression are normally both clear cut and
consistent. It does not reckon with the facts that such words
as "good" are, as Stevenson has reminded us, exceedingly
flexible and vague (that indeed is their merit), and that al-
though we may give them as much precision as is
in context
necessary for particular practical purposes, the general rules
governing their uses are entirely permissive in this regard. In
short, while we may contextually reduce the vagueness of
"right" or "good" for purposes of discussion, we may also, if
we so wish, avail ourselves of that very vagueness in order to
say certain very big or even sublime things with respect to
which it is perfectly appropriate. Limiting questions are
vague no doubt, but it does not follow from this that they
make no literal or rational sense. But even if they make no
literal sense, it still would not follow that they make no
sort of linguistic sense or that there can be no discursive
reason to ask them. For taking our ordinary language as a
whole, the rule is not that we must always speak literally or
that we must always remain well within the bounds of our
commonsensical applications of words, but, on the contrary,
that we may, by analogical and metaphorical extension, press
them into use in order to say many things of which common
sense may not approve and of which normal, literal-minded
people, as it happens, have no reason to speak.
The fact is that the great talk of poets, moralists, divines,
and philosophers is nearly always abnormal in some respect.
It is intended, however, not necessarily for abnormal persons
who have lost their minds, but for normal persons in those
outsized circumstances in which both language and every
other human resource are put on the stretch. Nietzsche said
that God is dead. Now I can well imagine that a common-
sense, ordinary-language theologian of Toulmin's stripe might
reply that since, by definition, God is eternal, Nietzsche's
utterance transcends the bounds of proper theological dis-

course and hence makes no literal, rational sense. Yet we all
do understand Nietzsche, and, what is more important, we
also understand him to be saying something of profound in-
terest and importance for our culture. There is, of course, a
sense in which his remark is paradoxical, but just because of
this it conveys the point Nietzsche wished to make more
poignantly and powerfully than a dozen pages of careful,
literal discussion of the atrophy of traditional religion.
I am not suggesting that it is impossible to talk sheer non-
sense. If I should say that the square-root of minus blue is a
close friend of my mother, I would be talking gibberish, if
only to prove a point. Nor am I denying that there may be an

important philosophical point in saying that statements of a
certain sort are nonsensical, even when with some ingenuity
we might find a small logical needle in the linguistic hay-
stack. A
good deal of metaphysical discourse, although not
gibberish, says obscurely and portentously what is hardly
worth saying when we boil it down. And the same is true of
much "philosophical" ethics which merely tells us in an elab-
orate, roundabout way that on the whole honesty is the best
policy or that those who ignore their obligations may come to
grief. But not all of it is so. When Kierkegaard talks of "sus-
pending the ethical" he is saying something which I find it
necessary to construe, but whether or not in the end I am
prepared to follow him in this regard, it is plain to me that he
raises a profoundly important question which I cannot ignore
and which I am bound to try to answer. The trouble with
Toulmin's philosophy of language, it seems to me, is that,
while it admits that we can ask limiting questions, it apolo-
gizes for and hence downgrades them in advance by making
them appear to be, strictly and "rationally" speaking, mean-
ingless. In so doing it ironically creates a contempt both for
ordinary language and for reason, which is precisely the op-
posite of its own intention. It seeks to protect our work-a-day
by throwing over it a cloak of built-in
ethical establishment
linguistic and which no morally well-
logical respectability
spoken person can wish to tear asunder. It succeeds, how-
22o Reason and Conduct
ever, only in disposing moral "outlaws," like Nietzsche, to go

beyond good and evil and to abandon all pretense to coher-
ent, rational criticism.
ON THE MEANINGS OF ETHICAL AND MORAL

This brings me to certain questions about the adequacy of
Toulmin's account of the use of the language of morals in
more ordinary contexts. One such doubt concerns his delimi-
tation of the logical use of the words "ethical" and "moral" as
applied to judgment. According to him, the function of an eth-
ical judgment whaj if j c fhaf malgAg 11g r>all if "pfhipal" "ig
the fact that it is used to haaaaooize peoples notions ." 3
_, .
But mis is surely false, even on his own account. For he ad-
mits that there are at least two levels on which ethical judg-
ments are made: (a) the level on which we judge the Tight-
ness of particular actions which fall naturally under some
rule prescribed by our moral code; and (b) the level on
which we judge the merits of the social practices of which
such rules are the leading parts. On the first level, we are
normally precluded, according to Toulmin, from asking di-
rectly whether a particular action as such would increase
( social harmony so long as there is an appropriate covering
\rule which prescribes the action as a moral duty and so long

âs its performance does not involve us in a conflict of duties. 4
(in short, many, perhaps most, ethical judgments are re-
garded simply as straightforward rule-determined assertions
about the Tightness of particular actions, which, on Toulmin's
account, are commonly supported merely by properly classi-
fying the action as of a sort that the moral code prescribes as
—
a duty. With respect to them there need be indeed, if he is
right, there can be —
no thought on the part of the judge
about the tranquillizing effects either of his judgment itself
or of the action whose Tightness is in question. But even
when we turn to moral judgments concerning general social
practices, it does not appear at all obvious that such judg-
3 Ibid.,
p. 145.
4 Ibid., p. 146.
ments must always be made with a view to increasing social

harmony. On the contrary, it seems entirely possible to judge
ethically that, on the whole, a certain practice should be
kept intact even though its modification would reduce social
conflict, on the ground that the practice itself has such great
? mtxinsjic__value that it overrides any inconvenience or dis-
harmony resulting from its observance. Moreover, it is also
proper to judge ethically that a certain practice should be
modified, not because the practice as it stands makes for con-
flict and disharmony, but simply because it is unfair or be-
cause it unnecessarily restricts the freedom of its practition-

ers. In a word, while social harmony is, for most of us, an
important moral good, it by no means provides the only cri-

terion for all moral judgments, much less the definitive one.
Unless I am mistaken, Toulmin is attempting, unwittingly,
to write his own utilitarian standards of ethical propriety into
the very use of such expressions as "ethical judgment" and
"moral judgment." To this extent he himself is confusing a
question of usage with a question of morals, or, better, he is
confusing a question of common or ordinary usage with a

question of proper or correct usage which itself can only be
settled on moral grounds. On those grounds, however, there

is little to be said for his view. For, to put the point in the
bluntest possible way, to convert utilitarianism from a sub-

stantive moral philosophy, as Mill envisaged it to be, into a
semi-logical truth even about moral reasoning concerning
social practices would be at once to trivialize utilitarianism
and radically to constrict the area within which substantive
moral reasoning may occur. For my part, I would not, even if
I could, write my own moral convictions into the very lan-
guage of morals precisely because this would reduce my own

subsequent freedom as a moral critic and agent. Toulmin, in
effect, is trying to make the rules that govern the usage of
ethical expressions doown moral chores for him. To this
his
extent he is reduce the area of his own respon-
also trying to
sibility as a moral judge. I believe he cannot succeed; but if
he could, I would deplore the result, if for no other reason
than simply that it would convert his own moral life into a
kind of game. Where morality is thought of, as it sometimes

is, as a matter of "playing the game," immorality automati-
cally becomes a matter not of conscience but of taste, and

those whose tastes run in other directions are then at liberty
to play an entirely different sort of game, or else to play as
they please according to no rules at all.
Here we come within sight of a fundamental error in Toul-

miris analysis which is reflected in his facile tendency to
[equate the terms "j*\}\iraY' anrl "m oral" as app li ed tn such
Vthincrg p? jnrlgrnpntg rprj^ arguments, and practices. It is a
tendency, I may add which is none the less mistaken be-

cause of its among contemporary moral philoso-
prevalence
phers. 5 Now
need not be denied that in certain contexts,
it
the meanings of the words "ethical" and "moral" overlap in

such a way as to make it extremely difficult to distinguish
them. Nonetheless, each has distinctive connotations, more
clearly apparent in other contexts, the confusion of which
can lead only to radical misconceptions of the language of
morals.
Suppose, for example, that we were asked by a novice in
the use of ordinary language for a clear case of an ethical
code. One perfectly natural answer, I think, would be to
/ mention the traditional professional code of physicians em-
bodied in the Hippocratic oath. Suppose then that our novice
i
went on to ask whether that code is also to be called a moral

code. To this it is entirely possible, or even likely, that we
r^would say yes, since many of the rules embodied in the Hippo-
1 cratic oath are similar to, and may even appear as mere
corollaries of, certain moral principles which we ourselves
âccept. Suppose, however, that we were also asked for a
clear case of a moral code. Now I admit that we might be
tempted to reply by citing once more the Hippocratic oath.
Nevertheless, the temptation should be resisted. For while,
at least in one sense, there^ may be cle ar, para digm cases of
ethical codes there can be, at best, only exemplary instance s^
,
of «-B3flral codes^ singe mrHTrnrLjririlre.s rvny ml£ot conduct n

moral principle save the conscientious hplipf of itsadherent
5 In other essays, I am frequently guilty of this error.
that he himself ought to live by it. The obvious reason why

ethical codes are so commonly confused with moral codes is
that the former often serve also as moral principles for cer-
tain individuals who hold themselves personally responsible
to their demands and who therefore feel a sense of personal
guilt when they violate such codes. The moral consciences of
organization men seem largely preoccupied with the per-
formance of duties imposed upon them as functionaries of the
institutions of which they are members. But few men are,
morally, merely organization men. And even within societies
as well-organized and institutionalized as our own there may
be men, such as Thoreau, who refuse on principle to regard
themselves as categorically bound by such duties while at the
same time having very high standards of personal conduct.
Such standards, by hypothesis, are the rules of no ethical
code; if there are such things as "moral principles," these are
moral principles par excellence.
Now Toulmin's characterization of ethical codes and hence
of purely ethical reasoning, so far as it goes, is not far from
the mark. It does well enough, that is to say, as a general
description of those rules which define what is regarded as
"ethical behavior" among particular classes of men who arejl
organized for certain particular purposes and who accepts
such rules as a condition of class membership. It also pro-ji
vides a description of the codes that determine what is to
count asjgroper or good conduJc f""ni the various profession s,
in service organiza tions, an d insuch institutions as the f am -
ily, the state or the gpriook
, But it does not adequately de-
scribe those distinctivelymora l codes which we have in
mind when we speak of "personal morality." Such codes,
even when they overlap the codes of various groups to which
we may belong, are not as such, systems of mles whieh we
must â ccept as the memhersof an y group, including th e
societôf mor al agents "Forthere is no such society, andv
. \/
there no general code of morality^ to which every moral
is
agent is beholden, f. moral community? as distinct from any

i
formally organized society, is merely a group of like-minded

moral persons, each of whom is responsible to the principles
of the group only insofar as his own conscience permits. In a

word, no moral agent can be an agent of morality, and no
mere agent of morality, if such there were, could possibly
qualify as a moral agent.
However, the issue here goes beyond the verbal question
of whether, without violating ordinary usage, we can speak
of a general code of morality for the critique of social prac-
tice which by definition is binding upon every moral agent
simply as such. It is whether such a code, even if it were pos-
sible, is, in the sense of the term with which we were con-
cerned in the above remarks, morally binding upon us. For
in that sense, no rule can be a moral principle and no code of
"morality" can be a moral code unless the individual persons
who adopt it believe, for their part, that they ought to live by
lit. Thus even if there were no linguistic oddity in speaking
Iof a general code of morality, it would still be an entirely

lopen question whether I or anyone else really ought to ac-
cept its rules. In other words, as individual moral judges and
agents, we would still have to face the problem of deciding
whether we ought to regard ourselves as functionaries of that
public institution which some may call "morality." For us,
what they call "morality" seems hardly a form of morality at
all. And in saying this, we are testifying to a fundamental
fact about the linguistic behavior of the word "moral" and its
cognates.
Thus are we brought back, once more, to the 'limiting
question" of whether we ought to be moral. In one sense,
which is not Toulmin's, that question may well be senseless,
as Bradley long ago pointed out. But in Toulmin's sense,
as it turns out, the question may not be so very limiting
after all for anyone but that Golem of contemporary sociol-
ogy, "the organization man." So choice a spirit as Thoreau
would not have thought it so; for, like Kierkegaard's Abra-
ham, he had long since "suspended the ethical" so that he
might at last hear the small, still voice of his own conscience.
In conclusion, let me say a word about Toulmin's thesis
regarding the alleged independence of morals and religion.
As he puts it, "Where there is a good moral reason for choos-
ing one course of action rather than another, morality is not

to be contradicted by religion. Ethics provides the reasons
for choosing the right' course: religion helps us to put our
hearts into There is no more need for religion to compete
it.
with ethics on its own ground than with science on its: all
three have their hands full doing their own jobs without
poaching." 6 This point, Toulmin continues, "is sometimes ex-
pressed by saying, 'We believe God's will to be good, not be-
cause it happens to be His will, but because it is good'; and it
only reflects the difference between the functions of ethics
and religion. If an action were not right, it would not be
'God's will' that we should do it. Or again, if an action were
not right, it would not be for religion to make us feel like
doing it." 7 It must suffice here to say that the last telltale
sentence of the preceding quotation clearly shows how far
Toulmin is either from a sympathetic understanding of reli-
gious discourse or from a comprehensive grasp of its func-
tions. To the man who believes in God, his religion does not
simply make him "feel like doing" what he antecedently
and independently knows to be right. Indeed, he may not
recognize it to be right at all apart from the fact that his
God prescribes it. Toulmin's own morality, like that of Mill,
is plainly secularistic. And such a morality, despite the
Church, is plainly possible. The interesting consequence of
his view,however, is that it requires him to think both of
morals and of religion in terms of specialized "jobs of work,"
which men of God and many conscientious moralists would
consider merely ludicrous. There are some religious mystics,
for example, for whom religion has nothing whatever to do
with morals, not even in the secondary "backing up" role
which Toulmin assigns to it. For them we do not even come
within range of the religious life until the ethical life has
been completely transcended. There are also many others for
whom the "will of God" defines, in the practical sense of the
term, what is to be accepted as morally right and proper. For
6 Op. cit., p. 219.

7 Ibid., p. 220. For a more extended discussion of this issue, see my
essay IX, pp. 171-205.
them, it is quite true that if an action were not right, it
would not, of course, be "God's will" that we should do it,

but then, in their eyes the only reason for regarding any act
as right is that God himself has willed it.
Once seems to me, Toulmin makes the mistake

again, as it
of confusing his own settled attitudes toward morality and

religion with the ordinary rules that govern the usage of
moral and religious expressions. It also mistakes the proper
moral autonomy of persons for the illusory logical autonomy
of something he calls "ethics" and "religion." What some of
Toulmin's secularistic readers may not see is that the auton-
omy he offers them as agents of morality is a mess of pottage
which "frees" them from bondage to an authoritarian God or
church only in order to subject them to a system of supposed
linguistic rules which they can escape, if he is right, only by
following Nietzsche beyond the horizon of moral good and
evil.
[XI]
A New Defense
of Ethical Realism
The seemingly naive title of Professor E. W. Hall's new work

1
is both accurate and misleading. It correctly reflects the
author's unquestioning assumption that in some sense there
are values. Indeed, Hall writes almost as though the emotive
theory had never been conceived. Yet any prospective reader
who supposes from this that Hall merely rings the changes
on traditional ethical realism will be in for something of a
shock before he finishes the book. Hall agrees that from one
standpoint "value" is unanalyzable, and that value judgments
are irreducible. But he takes this to mean something quite
different from what Moore, for example, appears to have
understood by it. Hall rejects even the formal model of de-
scriptive sentences as a basis for understanding value judg-
ments. According to him the latter have a concealed syntax
which is different in kind from that of factual statements.
They are also subject to distinctive validating conditions.
Hall admittedly never succeeds in defining the difference
between fact and value. In his view this is not due merely to
his own philosophical limitations, for in the end, he believes,
the difference cannot be directly stated at all. The only alter-
native open to him, as an analytical philosopher, therefore, is
to try to expose, as it were, the distinctive nature of value

through a series of critical studies of other objectivistic theo-
1 E. W. Hall, What Is Value? An Essay in Philosophical Analysis
(London: Routledge and Kegan Paul, 1952).

ries. Criticism is thus an essential adjunct of Hall's method of

analysis. Without any attempt to characterize value as a
it
unique ontological category, co-ordinate with fact, would be

frustrated from the outset. By what might be called the tech-
nique of successively less misleading approximation, at each
stage of which still another, more subtle categorical confusion
is revealed, Hall seeks gradually to "induce" in the reader a
firmer grasp of the delicate and complex difference between
value predications and other forms of rational utterance.
The essential puzzle involved in the analysis of value arises
from the fact that although value judgments do not assert any
matter of fact, they somehow include an unasserted refer-
ence to fact, which is revealed in the locution "ought to
exist." Strictly speaking, the answer to the question "What is
value?" cannot be expressed in a correct language. Indeed,

the very question itself, which Hall thinks "so eminently
basic and eminently sensible," seems inevitably to mislead us
into an implicit yet fatal blurring of the distinction between
what is and what ought to be. Thus the dilemma. In a cor-
rect language, there is no answer to the question. The only

possible answers must be formulated in ordinary language
which invariably lands us in paradox. Yet the question still
persists.
According to Hall, the difficulty arises not from the lack of
an adequate vocabulary, but rather from the nature of value
and fact, which are irreducible but indefinable features not
only of the logical syntax of language but also of what there
really is.
The resemblance between Hall's point of view in this re-

gard and that of the early Wittgenstein is not at all coinci-
dental. It is in full view throughout the latter sections of the

book. However, I myself was rather more struck by the un-
anticipated affinity, which Hall himself appears not to notice,
between many of his characterizations of moral discourse and
corresponding views of certain recent neo-Wittgensteinian
analysts in England. This similarity is all the more striking in
that Hall regards the departure of the later Wittgenstein and
his followers from what may be called the linguistic idealism
a
A New Defense of Ethical Realism 229
and ontologism of the Tractatus as philosophically retro-

grade.
Let me try to make this affinity a bit more explicit. In so
doing, I may better succeed in characterizing Hall's own
position, which is a difficult one to try to summarize in a sig-
nificant way. Since Toulmin has given us the most well-
known statement of the point of view of the English school, I
will use his work as a basis for comparison. First of all, then,'\
ethical naturalism is not a live theoretical option for either I
writer. Yet there is a common based upon similar

rejection,
arguments, of ordinary objectivistic approaches which con-
ceive value as a property. In effect, they both hold Moore to
be right in what he denied, but wrong in what he asserted.
There is a common sense also of the theoretical futility of try-
ing to state precisely what the meaning of "value" is. Yet
neither of them is prepared to follow the emotivists in con-
cluding from this that value is a pseudo-concept. Neither
thinks that value judgments can be properly explained as
disguised forms of first-personal imperatives (though Hall
toys with the idea that imperatives may be disguised forms
of "normatives" ) Both writers insist upon a sharp distinction
.
between "psychological" questions concerning the causes that

may induce us to accept a moral judgment and the "logical"
question concerning its validity or legitimacy. There is a
plain sense, they believe, clearly reflected in the procedures
of ordinary ethical language, in which value judgments are
"objective" and hence justifiable by an appeal to rules. This
means that they reject Stevenson's psychologistic account of
"disagreement in attitude" as an adequate account of ethical
disagreement on its logical side. 2 It means also that there is
common agreement that the notion of a "logic" of valuation is
meaningful. Such a logic, moreover, is heid by both writers
to be fundamentally different in character from the ordinary
deductive logic of the books.
But there is still another, deeper likeness which is as cen-
tral as it is elusive. It consists in —how shall I put it? —
2 Hall himself has elsewhere given a searching criticism of Steven-
son's analysis of disagreements in attitude.
covert anti-intellectualism which is at the same time com-

mitted to the defense of valuation as a form of rational dis-
course. Or better, and perhaps still more darkly, there is a
common underlying conviction that the defense of rationality
and objectivity in ethics must go together with a rejection of
any formal identification of factual and value sentences. For
at the end of the road of intellectualism and descriptivism in
ethics, however sophisticated it may be, lies moral irrational-
—
ism and this whether the intellectualism be naturalistic or
non-naturalistic.
These remarks are not intended to detract from the origi-
nality of Hall's analysis. Quite the contrary. They are in-
tended rather to place it in relation to other contemporary
discussions of the problems of ethics. I am all the more im-
pressed by this wide area of agreement precisely because
there is so much difference between the respective writers on
more ultimate questions. In my judgment it is this area of
agreement which will provide the focal point for further
analysis on the part of analytical moral philosophers in the
decade ahead. It is veritably the new thing under the sun
which offers a plausible way out from the impasse of the
emotivist-descriptivist controversies of recent years. For this
reason, despite several serious reservations on specific points,
I am prepared to say that careful reading of Hall's book is
mandatory on the part of all philosophers interested in the
progress of ethical theory.
I will now mention one or two aspects of Hall's later analy-
ses which bear upon some of my own basic doubts about his
position. One of the most suggestive and provocative parts of
the book is the section in which he discusses the possibility
that lvalue ." like -tmtb m ay he. a semantical predic ate. In
this section there is an especially acute analysis of R. B.
Perry's interest theory of value which Hall interprets, by
deliberately stretching a point, as a land of precursor of the
semantical predicate theory. This interpretation requires that
l teres t" be construed
as referring not to an organic behav-
ior pattern,but rather to a distinctive referential aspect of
^symbols themselves. So construed, the "object" of interest,
which is so puzzling in Perry's own account, is interpreted

not as the potential terminus ad quern of such a pattern, but,
somehow, as the referent of a certain unique mode of desig-
nation.
Such an interpretation, if it could be carried through,
would have advantages even for Perry himself. In the first
place, it enables us to make more sense of Perry's insistent^

reiteration that value is a relation into which things of any
ontological status whatever may enter with interested sub-
1/
jects. It enables us better to understand how there can in
some sense be value even when the object of interest, which
is not the mediating act of expectation but its cognitive con-
tent, is never realized at all. And, since interest is distin-
guishable from cognition, it makes it possible to preserve in
semantical terms, as a purely behavioral interpretation can-
not do, a formal distinction between value and fact. For if
Hall's interpretation be accepted, fact becomes the reference
to a cognitive or descriptive sentence, while value is the
designatum of a different sort of sentence which Hall calls,
for purposes of contrast, an "interest" or "value" sentence.

(This is not to be confused with Perry's own "interest judg-
3
ment" which is purely factual. ) But at the same time, since
every interest contains an implicit interest judgment, every
value sentence, on this interpretation, would include an un-
asserted reference to fact. This would be advantageous since,
if Hall is right, every value judgment asserts that something
ought to exist.
Hall recognizes that such an interpretation is incompatible

with many other things that Perry says. Above all it is incom-
patible with view that although "mediating"
the latter's
interest-judgments may be mistaken, interest as a whole is
never so. For every interest, in Perry's view, there is a corres-
ponding fact. This renders it impossible to say strictly that
any interest or, to adopt Hall's new way of putting the mat-
ter, that any "value sentence" is in error. One presumably
could say this, however, if Hall's interpretation were

3 By "interest judgment" Perry means any mediating judgment which
locates, identifies, or characterizes the object in which interest is taken.
adopted, although it is unfortunately still unclear, even so, as

to just what the validating conditions of such a sentence
might conceivably be.
In order to resolve this difficulty, Hall examines similarly
reinterpreted versions of Ewing's view that value is the
"fittingness" of an attitude to its object, and Brentano's neg-
lected but suggestive view that value is the being of an ob-
ject of a right love. In the end, however, he feels compelled
to reject such views, even when interpreted as holding that
value is a semantical predicate, precisely because of the fact,
as he reluctantly admits, that in ordinary discourse words
such as "good" are ascribed not merely to sentences but to
things. If "x is good" is essentially an elliptical semantical
sentence like "x is true," then it cannot be directly applied to
the extralinguistic world. Such a limitation, Hall thinks, is
fatal. And in order to overcome it, he tentatively adopts the
position that there are "zero-level" value sentences which,
however, are not true or false, but legitimate or illegitimate.
"X is good" thus remains strictly irreducible and unanalyz-
able, but we can now also say, by analogy with Tarski's fa-
mous condition for any definition of truth, that "X is good" is
legitimate if and only if X is good. The semantical term is
thus "legitimate" rather than "good" itself. "Good" remains a
term in the object language.
Here, however, a curious difficulty arises. All along Hall has
seemed to hold, as against theories which treat "good" as a
property, that the grammatical form "x is good" is logically
misleading, and that properly interpreted it is an elliptical
way of saying "It ought to be the case that A is xyz." It is
this interpretation also which appears to substantiate Hall's
rather obscure view, reaffirmed on several occasions, that
value judgments contain an unasserted reference to exist-
ence. Now I am not sure that I clearly understand what
references to existence are, whether asserted or unasserted.
But so far as I do understand it, I confess that I do not see
how Hall's translation of "x is good" contains either an im-
plicit or an what there is. Or if it does,
explicit reference to I
should then like to know whether Hall could produce a
declarative sentence of any sort which does not contain such

a reference. More important, however, I fail to see how one
can at once hold that "good" is a term applying to nonlin-
guistic entities and that "x is good" is an imprecise and mis-
leading way of saying that "it ought to be the case that A is
xyz." For this translation surely does seem to convert "ought,"
the word which now does the work of evaluation, into an ex-
plicit semantical predicate which ranges, not over nonlin-
guistic existences,but rather over statements of the form "A
isxyz" This point is clinched, I think, when we observe that
the sense of "it ought to be the case that. ." is itself very . .
close in meaning to "it is good that. ."

. .
Let me put the matter now in more summary terms. I do

not see how "good" can be a term in the object language if
"ought" is a term which ranges only over statements. On the
other hand, it does not seem to me at all obvious that the
term "ought" itself ranges only over statements. At the intui-
tive level, certainly, it is predicated not only of statements,
but also, like "good" itself, and persons. When I say
of actions
that a certain action ought not to have been performed,
clearly I am saying something about the action itself, and
when I say that a certain person ought to do something, I am
just as clearly saying something about him. Here, it seems to
me that Hall's tortuous "categorial" analysis raises more ques-
tions than it appears to solve, and obscures as many points
about ordinary moral discourse as it pretends to illuminate.
This point related to another of even greater impor-
is
tance. Hall never content merely to explicate the uses of

is
expressions. Like the early Wittgenstein he is impelled also

to invest their uses with a portentous power of ontological
illumination, so that when we have unpacked the meaning of
a form of words we have also disclosed at the same time an
objective or categorial trait of being. The assumption is that
forms of words are logical pictures which present the under-
lying structure of things in themselves and that if we wish to
understand clearly what that structure is we must reduce
our form of words to some basic standard forms of expression
which are themselves irreducible. As Wittgenstein himself
later came to see, this gambit is completely misdirected. Out-

side the Bible, language is not a means of revelation. Nor are
there any forms of expression which are logically primitive in
any absolute sense or which present the facts more perspicu-
ously than all the rest. Linguistically and logically, we are
always in the middle of things, and this being so, there is just
no way whatever to proceed to the fringe which uniquely
and unassailably mirrors reality. The aim of analysis is in-
deed to make us understand our ideas more clearly. But
this understanding comes mainly to an ability to handle
them more adeptly in doing the jobs of work for which they
are fitted. Knowing clearly what an expression means is, at
bottom, knowing how to use it correctly, and knowing how to
use it correctly is not so much a matter of "seeing" or "per-
ceiving" something but of being able to do something in a
certain preferred way. In short, sentences are not pictures of
any sort, but utterances which we make, either to ourselves
,
or others, in order to get something done. Or, to vary the
word pictures which they provide are not photo-
figure, the
j
graphs or x-rays which copy reality but compositions, ar-
\ rangements, forms of artifice, which we ourselves make in
\ order to accomplish our own ends. And if the reply is now
made that one such end may be simply to copy or mirror
reality, then my rejoinder is that this largely misses the point.
For making all allowances for the metaphors involved, it still
remains a question whether a particular expression actually

does serve the purpose of a copy and whether what it mir-
rors is anything more than the face that looks into it. In his
later years Wittgenstein was apparently fond of saying that
forms of words are and reflect forms of life. And this dictum
seems to me to contain a profound insight. But its value is
not so much ontological as sociological, not so much cosmo-
logical as, in the widest sense, personal and moral. No doubt
the language of valuation reveals something. But, in my
opinion, what it reveals is not a distinctive category of being,
but a form of action and behavior. Or if it reveals a category
of being, then we must completely revise our notions con-
cerning the nature of ontology. For then ontology becomes a

study of the forms of human existence — or conduct.
Thus, although I am sympathetic to the views which,
rightly or wrongly, I seem to find Hall sharing in common
with some of the younger English ethicists, I have grave
doubts precisely at the point where he goes beyond them to
—
say what he calls "the philosophical thing" not that I have
anything against philosophical things as such. In particular, I
find the whole attempt to derive metaphysical categories
from an exposure of "the" structure of human discourse not
only a hazardous but an essentially wrong-headed venture.
Here, if anywhere, is the quintessence of the ontological fal-
lacy. To my mind, there is something essentially "queer" and
paradoxical in locutions such as "there are particulars," "there
are states of affairs," "there are properties,"
—"there are values." and —in Hall's
sense wish to deny them. It
It isn't that I
is rather that I can't get the hang of them. I just don't know
what sort of thing to say in reply to the question "Are there

particulars?" It is like the embarrassment I feel when my
young daughter asks "silly questions." I don't know how to
answer the question, because I do not understand what sort
of question is being asked. Perhaps all the ethical realist
wishes to assert in the end is that valuative discourse is, in
some sense, rational and objective, and that one can make
significant mistakes in one's moral judgments. With this I
agree. But if this is what he is saying he goes about saying it
in a very odd and misleading way.
Hall evidently refuses to take the emotive theory seriously
because it is incompatible with the objectivity which he
rightly assigns to value judgments. Yet the whole tenor of his
analysis implies that there is not merely a substantive but
also a formal difference between factual and value sentences.
For him, value sentences do not in any sense describe the
facts of life; they are neither true nor false; their underlying
syntax, if not unique, is at least very different from that of
factual assertions. He strongly suggests also that whereas
declarative sentences express beliefs, value sentences may
be thought of and that this amounts not

as "normatives,"
to a difference inwhat they refer to but in the very nature
of the way in which they "refer." He talks, indeed, as though
there were "modes" of reference, in much the same way that
other analysts now talk about "modes" of meaning. But
precisely because "referring" is ordinarily thought of as ap-
plying only to the terms of factual sentences, Hall appears
to be faced with a strange dilemma: Either evaluative ex-
pressions strictly do refer, in which case, despite his pro-
testations to the contrary, he has in the end assimilated value
judgments to statements of fact; or they do not refer, in
which case all the talk about the peculiar kind of reference
involved in value judgments is simply a misleading way of
saying that their mode of meaning is nondescriptive, even
perhaps, in the narrower sense in which that term is now
used, "noncognitive." I submit, however, that one may defend
the objectivity of moral judgments without supposing that
its key terms must be construed as "referring" to anything at
all. The issue is no longer cognitivism versus the emotive
theory. Indeed, Hall himself suggests how "normatives" may

be construed as having a syntax and a logic without reducing
them to statements which are factually true or false. In this
way he helps to show how value judgments may be properly
said to be objectively validated without implying from this
that their primary function is referential.
One final comment. Now, as I have already admitted,
there is a sense in which every moral philosopher who seeks
tounderstand rather than merely to reconstruct the language
of valuation must begin as a "realist." That is to say, what
Hall calls "value sentences" are used to make statements about
matters about which plain men constantly raise questions of
validity and even of truth. Men do say things are good or
right, just as they say that things are green or hot or pleasant,
and when they do other persons frequently raise ques-
tions, which are not in the least odd, about the truth, validity,
or correctness of such claims. These ordinary, common-sense
features of evaluative discourse become linguistic anomolies
when such statements are regarded merely as alternative,
disguised ways of issuing commands or orders. We may go

further: our sense of linguistic propriety is in no way
strained by remarks about "the moral facts of the situation"
or even by statements which assert that particular judgments
do not "correspond" to such facts. These features of evalua-
tive discourse demand explanation, and we must be grateful
to the intuitionists for calling our attention to them year in
and year out in the face of the prevailing tendency of the
so-called non-cognitivists to ignore them or else to explain
them away. I myself have also called attention to them, over
and over, in my criticisms of the emotive theory and the
imperative theory of ethics. But to call attention to some-
thing is not to explain it. And the "realism" to which it
commits us in this case, although essential, is nevertheless
pre-analytic. In the same way, the so-called correspondence
theory of truth, although unexceptionable when taken pre-
analytically, is simply useless and redundant when offered
in explanation of the nature of truth. Of course true propo-
sitionsmust correspond to the facts, whatever they are; the
question remains what it is to say this.
Fundamentally, then, I find Hall's analysis unilluminating.
It serves to reinforce pre-analytic convictions or habits of
speech, but it merely gives the illusion of explaining their
import. In effect, it tells us only that when we say that
something is good we really mean good and
to say that it is
that when we say this sincerely we do believe it to be a

fact. It reiterates without clarifying the by now obvious point
that in asserting that something is good we are talking about
something, though what we are saying about it is also very
different from what we say when we assert that it is heavy
or big. Hall is free to say, if he likes, that what is involved
here is a peculiar mode of reference. But this is not instruc-
tive. For now one has still to ask what reference in this or
any other mode actually comes to. I can of course "refer"
to colors, people, numbers, mythical heros, theories, and
values, and I use words in order to make such references.
The only interesting question, it seems to me, is what I am
doing in making them. Once this question is raised, however,
itbecomes immediately evident that in using words we are

not always making references. Referring is one important
"illocutionary" act, as Professor John Austin calls it, but it is

not the only one. Nor is it logically or linguistically more
fundamental than any other. It seems plain to me that
what I am doing when I commend something as good or
prescribe something as obligatory is not essentially "making
references": I am simply commending it or prescribing it. I
may also say, if I please, that in commending or prescribing a

thing I am "implicitly" making references to its goodness or
obligatoriness.But this gets us further and further off the
track and at the same time misleads us into supposing that
commendations, after all, must be nothing essentially but
references to queerish non-empirical entities.
Here, I think, the followers of the later Wittgenstein, as
well as some Oxford philosophers who profess not to follow
him, ask better and more helpful questions than the realist
Hall. Actually their questions are not particularly new.
Berkeley, Hume, and Kant asked some of them many gen-
erations ago. And, more recently, the pragmatists have turned
such questions into a veritable method of philosophical anal-
ysis. If I correctlyunderstand them, both Peirce and James
insisted that if we want to answer questions concerning the
"meaning" of our ideas and if we want to give clear, decidable
answers to them, we must stop trying to inspect the "es-
sences" to which our ideas or words presumably refer, and
observe what we normally do in trying to confirm or sustain
the propositions in which they occur. This does not imply
that the meaning of an expression is, in the sense in which
the positivists understood that phrase, its "method of verifi-
cation." It does mean, however, that language is first of all
a form of speech, and that the only clues we have as to the
forms of actions which we perform in speaking and thinking
with words are the various activities and operations that
are involved in carrying them through. Like all realists who
cleave to their realism in attempting to make their ideas
clear, Hall leaves even their points of reference themselves
ultimately obscure. In fact he makes of reference itself an
ultimate mystery and the "objects" referred to in value sen-

tences something beyond the power of human discourse to
illuminate.
In the end, then, it strikes me that although the logic of
Hall's own analysis forces him in spite of himself away from
the descriptivistic reductivism which so seriously handi-
capped analytical moral philosophy in the twenties and
thirties, he is still unable to enter the philosophical promised
land to which his own suggestive study so plainly points the
way.
[XII]
A Revival
of Ethical Naturalism
Those for whom the life of reason is still an essential part

ofwhat should be meant by "the human condition" will find
encouragement and refreshment in Philip Rice's distin-
guished book. 1 It should also produce a certain amount of
static in the ears of those fair-weather rationalists who have
been beguiled by the angel voices that are again calling
from reason's backyard. Rice is one unrepentant son of the
Enlightenment who has managed to remain loyal to its ideals
without in the least blinking the fact that on this side of
Paradise life can be very real and very earnest indeed. In
him the truly perennial philosophy, which Mill called the
philosophy of experience, has found another powerful ex-
emplar and advocate. In these jittery days of existentialism,
neo-orthodoxy, and public philosophy, this very fact is an
intellectual and moral event of some significance. It should
help to dispel the prevailing illusion that seriousness and
intelligence are sworn enemies.
The biblical allusion in Rice's title is not accidental. He
underlines its point in his fine opening commentary upon the
ambiguities of the archetypal Genesis story; what follows
thereafter is, in one sense, merely an elaborate meditation
upon its underlying meaning. So to have conceived his sub-
ject was a stroke of genius that makes clearer than a hundred
1 Philip Blair Rice, On the Knowledge of Good and Evil ( New

York: Random House, 1955).
A Revival of Ethical Naturalism 241
pages of apologetics the essential continuity of analytical

ethics with the hoariest traditions of Western ethical re-
flection.This is Rice's lie direct to those opponents of philo-
sophical analysis who would have us believe that "meta-
ethics" is a frivolous game of words, played only for the
laughs by rocking-chair linguists who have turned their backs
upon the ultimate concerns of humankind. The logical anal-
ysis of moral discourse, as Rice conceives it, is a serious
and exacting business. It is, in truth, only a more careful
endeavor to answer questions about ethics that reflective
men have been raising ever since Adam and his wife left
Eden. To see such questions as, in part, questions about
language is not thereby to dissociate them from the moral
life, but to view morality, more accurately, as the primary
symbolic form of social deliberation. Forms of words, said

Wittgenstein, are forms of life. But if this is so, then anyone
who is seriously concerned with the latter cannot be in-
different to the former. It was because he wished to know
on what terms he must live and act that Rice was com-
pelled to make inquiry into the nature and efficacy of moral
notions.
Rice's point of view is plainly incompatible with that of
the self-acclaimed disciples of Edmund Burke who yearn
impossibly for a golden age of custom before sophists had
taught the classes to mouth sceptical questions about the
meanings of good and evil. These gentlemen regard them-
selves as realists. But the notion of a purely customary
morality, untroubled by naughty questions about its own
validity, belongs to the same limbo of philosophical and theo-
logical fictions that is inhabited by progress, the philosopher
king, and original sin. Since Adam, ours has always been
an open or reflective morality. Because of this, we have never
really been able, even when we have been willing, to let
Somebody Else decide for us what good and evil are. And
because of this, we have never for long succeeded in avoiding
or, as they are now called, analytical questions
the dialectical
about the meanings of goodness and moral obligation. The
only serious question to be asked regarding any moral
philosopher, therefore, is not whether he is engaged or
whether the object of his affection is virtuous, but whether

he has illuminated his subject and whether he speaks the
truth. It is easy to become engaged; it is not hard to engage
oneself to Virtue; the difficult problem is to understand why
Virtue is so called.
But Rice's book provides a more eloquent defense of
philosophical analysis than I can give. The only thing to say to
those traditionalists and obscurantists who prefer conformity
and commitment any price to understanding is simply
at
that it is better to be a Socrates, however unhappy, than a
contented cow that chews its cud in accordance with some-
thing called Natural Law. It may do better to forestall another
line of criticism which may come from some of Rice's own
brothers of the more immaculate conception. Rice turned to
the problems of analytical ethics, not because he was diverted
from his proper vocation as a moralist, but in order to ful-
fill that vocation more adequately. Meta-ethics remained for
him a means; his end, like that of Aristotle, was to live well.
This conception of the function of ethical theory is not
shared by all analysts, many of whom, unfortunately, have

transformed a necessary distinction between the use and the
mention of words into a pernicious bifurcation between
human practice and philosophical theory. In doing this, they
have also done the home-work of the firing-line philosophers
for them.
Rice does not blur the distinction between use and men-
tion. But he is aware, as some analysts are not, of the feed-
back, the effect, that is, of talking about his talk upon the
talking animal. An excellent example of this phenomenon
has lately been provided us by Walter Lippmann, the well-
known public philosopher. Mr. Lippmann is not the most
meticulous analyst of his age, but he has very strong con-
victions about the nature of good and evil and the source
of the moral authority. His meta-ethics, for the most part,
is a somewhat attenuated version of the doctrine of natural
law which Locke inherited through Thomas Aquinas from the
Stoics. What is significant, however, is not so much the
question whether that doctrine will bear scrutiny. That

question may have been settled by Hume in the Third Part
of his Treatise. More important is the fact that Lippmann
uses the theory as a prop for his own highly conservative
ideal of "civility" and his anti-utilitarian conception of the
whole duty of man. But Lippmann is not the only philosopher
whose feed-back has been working overtime. The same is
true of the followers of Bentham and Mill, whose intended
reforms in the theory of morals and legislation merely had
the effect of determining the practice of liberal moralists for
over a century and a half.
In short, theories of practical reason invariably turn out to
be something more than diagnoses of moral discourse. They
are, also, willy-nilly, reconstructions of second-level attitudes
which govern the actual validation and justification of moral U
judgments. No doubt they sometimes contain their grains
of truth; in that event, they can tell us something important
about ourselves that we ought to know. But they have a
strange way making themselves come true, ust to the ex-
of j
tent that^we take them as models of propriety for practical

reasoning. There is nothing mysterious about this. hlverything,
we have been told, is what it is and not another

thing. But
it is on occasion, that things turn out to be what
also true,
they seem. A rose is a rose; it only needs to be. But we,
it happens, are not roses. We fancy ourselves to be creatures
of a certain sort, and then fashion ourselves accordingly; or

else we imagine that we have a certain nature and destiny
and then conspire against destiny in order to prove that
we have no nature. In saying this, I should not wish to give
more comfort than necessary Like the
to the existentialists.
rose, man also has a nature of sorts, but it is mainly an
acquired characteristic which we impose through our talk
upon one another. Whether the images of man that Freud,
Jung, and Marx have created bear much resemblance to their
original is, as we say, a nice question. Meanwhile, the rest of
us have been helping them out by taking on an Id, a racial
unconscious, and even membership in one of the social classes.
It was for just this reason, as I remember, that Plato thought
it undesirable that the truth about man should get around.

between
Rice's constant attention to the relations holding
methods employed in the study of moral ideas and those
that are used in practical deliberation is the source of many
of his most searching criticisms of such opposing theories of
ethics as intuitionism. He does not believe that there is any
evidence for the supposedly simple and "non-natural" quality
which G. E. Moore, the father of twentieth-century intui-
tionism in ethics, identifies with moral goodness. But his
more serious objection to intuitionism is that it employs a
method of analysis which makes for dogmatism and mystifi-
cation in the conduct of life. Bad methods of analysis
produce mistaken theories, and mistaken theories result,
one way or another, in bad practice. Actually, there is an
teresting parallel in this respect between Rice's position
and that of Moore himself. It was from Moore that most of us
ave learned the importance of distinguishing questions
about the meaning of goodness from those concerning the
things that in fact are good. But Moore distinguished these
questions because he believed that only the former are cor-
if
rectly answered can the foundations of practical moral

science be securely laid. The trouble with erroneous theories
of "good" and "evil," such as hedonism, is that they invariably
1 result in faulty deliberation, bad judgment, and, hence, wrong
action. By mistakenly identifying goodness with pleasure, the
1 hedonists commit the "naturalistic fallacy" of confusing the
\ pr operty of gnndnpss wit h snn^fhjngjglsg whicjl. maVbe liQm-
M- nmoiLia the things that are intrinsically goo cL More important,
however, they thereby are prevented from raising the en-
tirely rational question whether pleasure alone is really good.
It may turn out upon reflection, of course, that pleasure is
the only thing to which we can justifiably ascribe intrinsic

value. But even if this were so, it is still, for reasonable men,
an orjgn-efuestiorij^nd its answershould not be settled by
definition . in-advance^ of inquiry. In short, questions of~cTefi-
nition are, particularly in the domain of ethics, questions of
the utmost practical importance. Taken at its own word,
analytical hedonism makes the open question appear silly by
converting the principle, "The only intrinsic good is pleasure

and the avoidance of pain," into an apparently vacuous
tautology. In the same way, according to Moore, those meta-
physical ethicists who identify goodness with Being or Real-
ity at the same time renounce the moral right and duty to
ask seriously of anything that is, whether, really, it ought to
be. And because of this, theoretical errors on the part of
metaphysicians, from Augustine to Hegel, have been paid for
heavily in each case in terms of spiritual confusion and dis-
order.
Rice, the naturalist, and Moore, the intuitionist, are thus
both preoccupied with questions of method, precisely because
they are aware that methods of analysis in ethics powerfully
affect our basic attitudes toward moral deliberation itself.
Each in his own way seeks to understand the nature, condi-
tions,and limits of ethical justification in order that he may
think more clearly about the problems of conduct with
which they are both faced as human bein gs. Moore .opposes
naturalism hecausg h^ think g it deprives ns r> f the abHaty-to
raise s eriously the essential critical question^ "But is rJL after
•all^good?" Rice oppos e
jutuitioniaCQ because it makes a
mys tery of the property of goodness a nd because it thereby
r emoves moral judgment and jjplihpraiioD. from the -sphere.
1/
in which P"^ 3^ 0bservable eyiglfri''^ ran hp rpqiipsl-prl nr
1lV>1
,
_given . In this way, despite Moore's intentions, intuitionism
provides others with an escape into dogmatism and hence
into irrationalism. For even if its "insights" turn out to be
true, it has no of showing why they are so. Rice's own
way
rationalism, however, is not hypertrophied. He believes in
reason, but he has no illusions about the reality of the surds
of human existence. His only aim is to keep them from mul-
tiplying beyond necessity. The moral life, he is quite aware,
is not wholly encompassed by the magic word "rational." It
also has its non-rational dimensions of emotion and decision.

The problem is to keep them from becoming confused, as
Pascal and Rousseau confused them, with reason and knowl-
edge.
Although he had an instinct for system which is in evi-
)
dence throughout the book, Rice has not left us a fully

worked-out system of philosophy. It is necessary, therefore,
to reconstruct his views upon the larger questions of meta-
physics and theory of knowledge from brief passages that are
to be found in On the Knowledge of Good and Evil. But
his moral philosophy is consciously formulated as the part
and function of an inclusive world-view, and it is worked
out in the light of a rigorous methodology. Rice's basic
philosophy is broadly empirical in method and naturalistic,
but not materialistic, in ontology. (It should be pointed out,
in this connection, that Rice's ontological naturalism should
not be confused with his "naturalism" in ethics. The two are
consistent, but neither entails the other. The latter, as Rice
%
understands it, is simply the view that ethical judgments
sometimes contain factual information which can be veri-
fied by ordinary scientific methods, and hence that there are,
in context, empirically discernible characteristics that func-
i
tion as the "Conferring" or "Identifying" properties oL.Yah.ifl.
As a metaphysical naturalist, Rice believes that what we call
"nature" is a self-contained system of phenomena that oper-
ates according to immanent laws without miraculous intru-
sions from any supernatural realm outside it. As an em-
piricist, he is committed to the "method of experience," as he
calls it, reminiscently of Mill. For him, all justifiable beliefs,

whether in ethics or in any other sphere of knowledge, rest,
directly or indirectly, upon observation. However, these theses
are not held by Rice as unarguable dogmas. He arrived at
them, painfully, only after prolonged study of their major
alternatives. Even when, in the end, he could not accept
them as they stand, he strove to discover their underlying
rationale and to utilize what concealed truths they might
contain.
Rice's method and ontology were stern task-masters. They
permitted him no easy recourse beyond experience in order
to justify his conviction that there is knowledge of good and
evil. His results are all the more impressive, in fact, pre-
cisely because of this. Since his findings do not depend upon
the disclosures of a special faculty of intuition or insight,
others may verify his findings, at their leisure, by garden-

variety methods that should be accessible toany person who
is willing to use his headpiece. And because he does not
claim privileged access to the revelation of a supernatural
authority, it is not necessary to undergo a conversion in
order to agree with him.
naturalist in ethics, Rice has a much tougher job on
As a
hishands than did his predecessors. It scarcely occurred to
John Dewey or to R. B. Perry, perhaps the two most dis-
tinguished of the older naturalists in this country, that the
search for definitions of "good" and "ought" might come to
nothing more, than the conclusion that these words
finally,
'
are^not terms of knowledge at all. Apail fiuni iiili annual
disputes regarding the particular empirical property with
which good or value is to be identified, the older naturalists
had nothing more to contend with, or so they supposed,
than the accursed intuitionists with their everlasting open
question, "But is it good?" But what if there is no property of
goodness, and what if "good" literally designates nothing at
all? Since Hume, naturalists and intuitionists alike have been
puzzled by the nature of the relation between the "ought" and
the "is"; but perhaps the difficulty does not lie in anything
more recondite than the judgments of value and\
fact that
obligation are not statements of any sort, but are merely
expressions of emotion. In short, what if there is no suchj
thing as the knowledge of good and evil, and if moral dis-^
course is merely a highly stylized device for venting and,
irrigating our attitudes?
When Rice himself first began to write on the theory of
value, such questions as these had not been seriously con-
sidered by most moral philosophers in this country. His own
earlier papers, which appeared intermittently in the pro-
fessional journals for nearly a score of years, reflected the
then prevailing assumption that there is an objective prop-
erty of value, the only serious theoretical problem being to
define it however, this problem has
correctly. In recent years,
been found be egregiously question-begging. Advanced
to
discussion, therefore, has shifted radically from the ancient
debate between the naturalists and intuitionists over the

definability of "good" to a new and far more crucial argu-
/*ment between the "cognitivists" or "descriptivists," who con-
l tinue to believe that there is a property of goodness, and the
y'yemotivists" who hold that "good" is not a descriptive term

(and that value judgments have a merely expressive and
/ incitive function. Rice himself remains a cognitivist, but with
a vital difference. He still believes that there are ethical
beliefs as well as ethical attitudes, and that ethical proposi-
tions are empirically verifiable. But he now faces head-on,
with no lingering trace of question-begging, the challenge of
the emotive theory. Because of this, his own naturalistic or
quasi-naturalistic moral philosophy has been immeasurably
strengthened.
Bice is a skillful strategist. Realizing the essential elements
of truth implicit in the emotive theory, he very wisely prefers
to lose the opening battle in order to win the final campaign.
He acknowledges froni^the outset, therefore, that what he
calls theyjnatrix meaning) of ethical terms is non-cognitive.
The primordial common role of "ought" and "good," which
he refers to, suggestively, as their "trigger function," is not
that of describing the character of an act or of its result,
but rather that of expressing the fact that a decision h as
been reach ed or o£^sjgnalhng the release of a contemplate d
^ line of action. No emotivist has described more carefully this
peiformatafe di mension nfrnoral discourse, and none has ex-
plained more clearly why iF cannot be reduced without re-
mainder to the descriptive dimension.
However, there are pervasive differences between moral
judgments that prescribe what we ought to do and simple
first-personal imperatives that tell us, without ado, to per-
form an act. In clarifying them, Rice already shows how far
rxgmoved is from the cruder forms of emotivism.
his theory
I In the first place, moral judgments are expressed, as a rule,
I with a characteristic tone or accent which is at once de-
/ A liberate and judicial. Reflecting upon this, we are made to
i think at once, not of someone giving orders, but more of a
I
judge instructing the jury or delivering a sentence after the
verdict. Thepoint is that in moving from the simple impera-

tive to the moral judgment, we are aware, so to say, of a
shift in "voice." In the former case, we speak to someone
for ourselves and on our own authority; in the latter, we
speak more impersonally, and the fact that it is we, Stan
Spatz or Joe Doakes, who happen to voice the judgment is
I
y
felt to be irrelevant. The l anguage of morals enables us to
voice rlninr; th nf rtr r tnl rr-n by both parties to he "objectively"
binding uppn a11 moral agents There are also tell-tale
grammatical and syntactical differences between moral
judgments and ordinary commands. Commands are ex-
pressed, characteristically, in the imperative mood, whereas
ethical discourse employs the indicative. For this reason it is
nonsense to inquire about the truth of "Pass the cigars," but
entirely natural and proper to ask whether "Pleasure is the
only intrinsic good" is true. It is no accident that generations
of moral philosophers, however mistakenly, have adopted
the model of factual statement in analysing the meaning of *\
moral judgments. For if these grammatical and apparently J

logical features of moral judgment make no difference, at/
bottom, to what is asserted through them, why then should^
they exist? W ere such judgments merely oonoealed impera-J
rives, moral diseonrse could be eliminated without serious
Joss,,, In that case, one would suppose, it should be eliminated,
since it serves only to mislead not only philosophers, but \

even ordinary men, as to what is being said. In short, the \
emotive theory makes of moral discourse a linguistic anomaly, I
at least for those who accept the sensible principle of analy-

J
sis that natural languages do not, in general, multiply forms
of words beyond necessity. Surely, then, the heavy burden
of proof rests with those who hold, despite all this, that
the language of morals is essentially the same in meaning
and function as that of imperatives.
Rice thus contends that the grammatical aspects of moral
discourse are evidence of its distinctive role within the
wider domain of practical discourse which also includes com- -¥
mands, requests, invitations, oaths, blessings, and the thou-
sand and one other special ways of doing things with words.
This role is distinctive, besides, in at least two fundamental
respects, one of them having to do with the sorts of attitudes

which moral discourse expresses or evokes, the oth er havin g to
•do—ttrith the mediating Cc^gnl5v^~s et^which it releas es. As
we have just seen, the attitudes incited through moral judg-
fl^ ment are not those of immediate, unreflective compliance,
but rather those of impartial scrutiny and deliberation. The
conclusion to a train of moral reasoning is expressed in the
/"form of a judgment, not of a decision or demand. To say
I that an act ought to be performed is by implication to
\ vouch for it, impersonally, as worthy of performance on the
part of any man. Indeed, this is the feature of moral dis-

course which Kant, who himself is responsible for confusing
moral judgments with imperatives, so clearly understood, and
it was for this reason that he could speak of judgments of
moral obligation as rational, while insisting on an absolute

distinction between pure speculative and practical reason.
But if ethical terms function as triggers of a certain sort,
they are not merely triggers, but also sights. In this fact
Rice locates the possibility of an unPickwickian knowledge
of good and evil. By itself, as he says, the matrix meaning
is empty and blind; it tells us to go or to be ready to go,
but not where to go or how. It serves to prescribe, yet

unanswered the all-important question as to the
leaves
\p, ^characteristics thatan act should have if it is properly to
v> be prescribed as right or obligatory. But how are such char-
J
acteristics to be identified? In Bice's opinion, they are to b e

/found, in thn firtf n nt nnrn^ i n the n rHinary lower-level efh jeal
i
(
m axims which supply our conventional c tand grr nf ^'g^<-* <'
V and wrong. These criteria are not explicitly mentioned in

particular moral judgments, and unless some issue is raised
concerning the Tightness of a particular act, they remain
implicit. The fact that they are there becomes clear only
when the request for a justification is made and answered.
Now if the emotive theory were correct, then anything that
happens to set off the trigger of ethical terms could properly
be said to justify our doing what moral judgments prescribe.
With such a view, in which the connection between an
A Revival of Ethical Naturalism £5!
ethical judgment and its supporting "reasons" is seen as

entirely psychological, hard to see how it could be denied
it is
that if "All camels have two humps" caused you to agree that
you ought to keep your promise, it would be as good a
reason as any other. In practice, however, the justification
of moral judgments is not so wildly open-textured as this.
Not just anything that we may happen to say about an act is
,
considered relevant in answering the question, "What fea-
|
tures of a right act make it right?".Ji!ome characteristics of
\ acâswe, think janlm fhnm rinrH qj h a ir essentially up on
1 their rightness: othe rspl ainly do not. But how can this b e,
unlessJmplicit in the very use of "right" are criteria that, i n
context, restrict its range of application?
Here is the crux of the matter , and in answering this
question Rice seeks to make good his claim that, in addition
to their distinctive matrix meaning, ethical terms also have^-l
a meaning which may properly be called "cognitive" or 1
^
"descriptive." For ifther e are prevailing, interpersonal stand- /
ards^that goyern th^ ^pphrnt i nn T f~Wa1"gHvp terms
t r
particular contex fsT so that in siich^gon texts these_standarjds

are normally- part of what-isqinderstood when the terms ar
us ed, not implausible to say_ that they are p art of *hp
it is
ver^rasapiiig nf the terms themselv es. Consider, for example,

the use of "good" as applied to poetry. If the emotive theory
were then anything we might say about a particular
true,
poem could properly be called a reason for regarding it as
a good poem. But this is plainly belied by the common
practice of literary criticism. The fact that a poem happens
to express desirable political or theological sentiments does
not, as we say, make it good poem. And if some benighted
a
critic should call it a good poem for such a reason he
would be regarded as incompetent. The "as we say," here,
is essential. What it suggests is that when applied in a literaryV
context, "good" is ordinarily understood to have a very lim- V
ited range of application. This range of application, in turn, / y

presupposes a built-in standard of merit for literary works
j
which is quite different from those standards by which we J
judge social policies or submarines. Such qualifying adjec-

""
/fives as "aesthetic" owe their use precisely to the fact that
( they serve to render explicit a frame of reference that is
^ already tacitly supposed, in the literary context, in the use
C of "good" itself. Where there are such (gtaridard~contextSg
although "good" functions as a trigger, it also functions as
a sight. Thus —although "good" is a term of praise o r com-
mendation, it also serves, in norma l oirrnmctinrp^ tn ffiwng
our attention upon certain characteristics that are expected
and'Tgquiied uf thing s so commended._
How such criteria have been acquired or modified is
another question. ea sy atjhis point, to confuse questions

It is
of m eaning or validity with questions of_orjgin. Although

Rice is at times sorely tempted to justify certain validating
standards on what would appear to be essentially psycho-
logical grounds, on the whole he keeps genetic and logical
issues reasonably distinct. But, again, a distinction is one
thing; a bifurcation is another. Standards are not written up
one way or another, by men,
in heaven; they are acquired,
and they are adhered to or broken for reasons that function,
psychologically, as causes. If a reason for doing something did
not in the least dispose us to do it, we would be hard put
to understand what might be meant in calling it a "reason*
/ at all. Because they wholly ignore questions of genesis, the
\
intuitionists are led, without knowing how or why, to talk
V of "objective" standards of right action. Likewise the pro-
ponents of natural law tell us that there are certain "self-
/ evident" rights and duties that are universally binding upon
^ / all rational agents; yet they leave utterly unexplained how
they are so, thereby permitting the skeptic, who happens for
1
\ the moment not to feel their moving-appeal, to deny the very

(meaningfulness of such expressions. In Bice's view, the pres-
ence within the language of moral^ of (^textual standa:^
A for the application of ethical terms reflects_certain_pxelinguis-
Hn âhitc that /^""T^ Pvplpjr^rl
vahiaHrvna"! only jn cnofrt-
jpsychologicalterm s, Such habits exist, but they are not

acquired at random. They are adopted, for the most part un-
consciously, in accordance with empirical laws of learning
and motivation that describe the formation and modification
»
A Revival of Ethical Naturalism (25$

V '
of human wants. Rice an environmentalist. But in an in-
is
teresting chapter, which he calls "The Appeal to Human

Nature," he speculates upon the possibility of a congenital or,
as he calls it, a "structural" a priori, which lies behind and
gives efficacy to our linguistic-conceptual a prioris. In any
case, whether there actually are certain built-in regulative
principles of human thought and action as Kant believed, or
it is just that we think them to be such because they have
become second nature to us through social conditioning, the

fact is that our maxims and principles are not bolts from the
blue but the artimilatio n of ps ychologic al dispositions to
thi nk and act .
Rice has many illuminating or suggestive things to say

about the genesis and mutation of value-standards. His un-
usual command of contemporary psychological and social
theory is in evidence throughout this part of the book, and it
gives a decent firmness to reflections upon the nature of man
that, in most other philosophers, tend to become impossibly
speculative. Nowhere does he permit himself to go very far
beyond the empirical evidence in the interests of theoretical
order or for the sake of a unified explanation of the human
tendencies which he behind our ethical standards and prin-
ciples. Above all, he refuses to succumb to the temptation,
which has overwhelmed so many psychologists and social
scientists these days, to convert a description of human
practices into a moral law or to derive a justification of\
moral universals from a theory of underlying similarities in \ ^/
the social customs of different peoples. For Rice, as for Mill )
before him, the logical derivation of practical principles trorn
theori es of man or of society is always illicit. Rice's "natural-
^
ism" in ethics is thus always qualifiecTby his^recognition of
the generic difference between statements about interest
and expressions of interest, and between descriptions of
human tendencies and the ethical prescriptions that are de-
signed to guide or correct them. The "appeal" to human na-
ture is of importance if what you want is an explanation of
how
your aim
standards are acquired or changed;
is to validate them.
it will not do, if
V
(25^) Reason and Conduct
Perhaps the most important feature of Rice's lengthy analy-

sisof moral justification is the distinction he draws between
A the tlgooaNmaking" characteristics., which confer value of a
I certain sort upon a particular class of things and those identi-
i/ I fying properties which, as he thinks, are "common to all

things or acts of which any normative term, such as 'intrin-
'
sically good' or 'morally obligatory,' can be asserted." In an
older philosophical tradition, the former were usually re-
ferred to as "vjrijigsl; Rice calls them ^Conferring Properties ."
The rules through which these properties are specified are
be found in the innumerable jnaxim s, formul as, and_grad-
in g_ schemes by^whi ch we regulate the assortment of prac-
ti ces and activities that Ims weT'to our special comm on m-
jerests. To list all of the Conferring Properties, or the rules

that define them,would be an endless and unprofitable task.
For what we would have, when we were done, would be
nothing more than a catalogue of the innumerable ^vayfc in
which things ma yb e_suitable _flr satisfactory to us. Ttwould
provide no illumination of the problems of valuation which
confront us when excellence competes with excellence or
when Nor does it shed
various virtues clamor for preferment.
on the question of what we are to do when it becomes
light
necessary to modify a whole grading system or an entire
/--technique of conventional commendation.
The question arises, therefore, whether there are any
interpersonal and super-ordinate principles of justification to
which appeal can be made "objectively" when it becomes
ed
necessary to decide between rules or when particular pro-
cedures require modification. Rice is not content to settle, as
some of his contemporarieshave done, for the pluralism im-

plicit in the job-lot of Conferring Properties. Nor does he
admit that with them we reach the limits of objectivity and
cognitive meaningfulness in the use of ethical terms. Beyond
them there are certain justifying principles of intrinsic value
to which appeal is made in justifying or criticizing the rules
through which the Conferring Properties are specified. The
characteristics defined or stipulated by such principles as
essential to anything worthy to be called "intrinsically valu-
A Revival of Ethical Naturalism (•55)
able" are called by Rice Îdentifying^ Properti es^* It is in the

discussion of them that one finds the closest tie to his earlier
work and through this to the older tradition of ethical natu-
ralism in this country.
For Rice, the fundamental problem regarding Identifying
Properties is whether, finally, they are one or many. After/
much detailed analysis, which cannot be adequately repre-

sented here, Rice eventually concludes that there is only one
such property. Following C. I. Lewis, who has greatly in-
fluenced Rice's thinking at this point, beholds that the basin
nnrrpative rategnry i ntrinsic value, apjDliejjonly within ex-
perience jiseJ£-{ "things"arie only extrinsi cally goo d}. He also
agrees with Lewis that the feature which is common to all
experiences that we would be willing to commend as in-
good is that "dimension-like mode" or range of
trinsically
feeling which we more or less inadequately designate by
c
such terms as "pleasure^ and enjoyment." It is important to
-!
G
a correct understanding of Rice's theory, however, that the
Identifying Property not be simply defined as pleasure. For
none of the ordinary feeling words in English covers the
entire mode which is here
or dimension of experience i:
question. For Rice, every pleasure is a case of intrinsic

goodness;^ but not every experience which exemplifies the
identifying Property would normally be spoken of as "pleas-
ant" or as "a pleasure." When we try to make the word
"pleasure" answer for the whole dimension, as the older he-
donists have done, we stretch it unmercifully beyond its
normal range of application. The result is the inevitable mis-
understanding that has attended such forms of hedonism as
Epicureanism and Utilitarianism.
Whether Rice's doctrine is classified as hedonistic is not
important. It is very any case, from those forms of
far, in
"naturalistic" hedonism which regard "good" and "pleasure"

as virtual synonyms. Nor is Rice's analysis, whatever its
other faults may be, open to any of the customary objections
that have been made, again and again, to the textbook
"
varieties of hedonism. Rice-^never asserts that good" Jk
identical in meaning with any ordi nary feelin^pvordLsucl^ as
6
(25^) Reason and Conduct
/^pleasure," "enjoyment," "delight," or "satisfaction." "Good"

Valways has a distinctive matrix m eanin g not possessed by any
O /of these terms. And beyond this, none of them, as it stands,
/conveys all of what is meant by the "Identifying Property"
ôf intrinsic value.
In his book Rice has become at once more sensitive to and
more respectful of the proprieties of ordinary language than
he was in his earlier papers, with the result that his whole
analysis is now vastly more subtle and far more adequate
to its subject. He is no longer tempted to cut Gordian knots
by arbitrary stipulation or definition, and because of this he
now rarely, if ever, misleads either himself or his reader
by what are only apparent "clarifications" of the concepts
whose meanings he seeks to understand. Most important of all,
he now sees with far greater clarity the logical force of the
arguments that have been raised against earlier hedonistic
theories of valuation, including even that of Lewis. Rice's new
respect for ordinary language indicates, I believe, an increas-
ing influence of recent Oxford philosophy upon his thinking
during his last years. This influence, I also believe, was
wholly benign, and perhaps it was greater than Rice him-
self fully realized. The arresting parallels between Rice's
theory of justification and those of such younger Englishmen
as Toulmin and Hare are not accidental. Yet it is equally
important to observe the points of difference between them.
In the first place, Rice is far more tenaciously systematic
in his thinking about the problems of ethics than are most
of the philosophers of ordinary language. He does his proper
share of detailed, piecemeal analysis in the book, but in the
end he aspires to give us a comprehensive theory of valua-
tion which will fit such analyses into an intelligible pattern
of inter-locking hypotheses. Moreover, while he respects
ordinary language he does not regard it as a symbolic para-
f
gon incapable of modification or improvement. As in the
V case of the Identifying Property of intrinsic value, ordinary
J
language may not happen to provide terms of sufficient gen-
* yerality and clarity for the jobs of work that philosophers want
(performed. In that case, it is necessary to resort to jargon
or stipulation if what one wants to say is to be said at all.

Finally, Rice finds the somewhat easy-going pluralism of
such writers as Toulmin, Hampshire, and, if I may say so,
myself, essentially unphilosophical. A nd just as he will not
rgst-wii]i_a_j)l urality of basi c pri nciples of intrinsic value, so
he refuses to accept the idea that there are peculiar rules of
inference within the domain of ethical reasoning. The logic
of morals, according to him, is no different from that of
ordinary deductive and inductive inference; only its terms
and the function of its principles and conclusions differ
from those of factual or scientific discourse. Moral principles
are not rules of inference, but premises of arguments. When
so regarded, the supposed need for a "third logic" of ethics
to cover the passage from factual premises to ethical con-
clusions entirely disappears.
It should not be inferred from this that Rice is unaware
of the margins of vagueness and uncertainty that are in-
herent in all moral deliberation and criticism. He likes things
to be as tidy as possible, but he is aware of Aristotle's
wisdom in saying that in the ethical domain things are true
"only for the most part." He believes that there is_only one
supremo principle of intrinsic val ue, and he thinks that it can
be given a vindication of sorts. Yet he knows that moral
principles, unlike stars, are not born but made, that their
guidance is uncertain, and that they do not automatically
apply themselves to particular cases. The guidance they offer!
for the conduct of life, when all is said, is hardly more >,
than that of a basic orientation toward the solution of prac/
tical problems. They are not recipes for ethical cookery or
doctor's prescriptions for what ails you. The philosopher, in
Rice's view, "cqnnot be finally convince d th^ê^ômam of
values is chaotic, that it will nltirnatejy resist his passion for,
finding or making intellectua lorder/j* But Rice does not
blink the difficulties that stand in the way of moral order.
Nor does he suppose that any particular philosopher or
school of philosophy has managed, once and for all, to
triangulate the good. His tentativeness, throughout the book,
2 Italics in the text.
is only one more evidence of his own intellectual discipline

and civility.
I, for my part, greatly, if also somewhat ruefully, admire

the systematic unity and constructive power of Rice's moral
philosophy. These are qualities that are all too rare in con-
temporary philosophical writing. Yet I would also be less than
candid if I did not say that I do not always find Rice's own
system wholly convincing as it stands. In one of the later
chapters he ar gues at length against my own irraclw>ih1f>
plurality of~e0tica1 prin m'p W
Tt has seemed to me that the
^principles of njbert^ and oCjusticeîmpose demands upon us

which sometimes conflict wrHTthose of the principles of least
^s ufferin g and greatest happiness Hence I have been unable,
.
in recent years, to accept the thesis that there is only one

Identifying Property which is essentially common to all cases
of intrinsic value. And I have offered certain "hard cases"
which are intended to show this. The tactics of Rice's
reply are sound enough. He argues, first of all, that the notions
of liberty and justice are not so clear as to enable one to be
quite certain, in particular cases, that the principles of liberty
and do make demands which conflict with those
justice really
and happiness. This is true, but perfect
of least suffering
/"clarity, in this sphere, is something which I, no more than
C Rice, aspire to. Vague as they undoubtedly are, the princi-
ples of liberty and justice express orientations toward social

^relations and individual life that have involved men of good
[
conscience in fundamental conflict with the principle which
^ he would have us accept as the principle of moral justification.
What are we, in conscience, to do? Rice's whole theory of

Conferring and Identifying Properties of value requires us to
accept as given those standards which are interpersonally rec-
ognized and invoked by men within a common culture. Any
appeal beyond them must of necessity be to individual
conscience or interest. At that point, however, we pass, in
,
r one sense, quite beyond the bounds of "rational" justifica-

.'
/V I tion to something else which can only be called personal or

' existential concern. If this is lacking, or is otherwise moti-
A Revival of Ethical Naturalism 25^
vated, the appeal, rational or otherwise, is simply defeated.
More bluntly, it becomes an apostrophe.
The issue here is fundamental, but must not be mis-
it
understood. Such conflicts of principle as I have mentioned
are not external conflicts between competing systems of

morals. They are not analogous, for example, to the con-
flicts which fundamental points, between Catholics^
arise, at
and Protestants or between Fascists and Liberals. Rather are <y I
they internal conflicts that occur within a common semi- J

system, the one, as I think, with which most of us, including
Rice, must try to reconcile our actions. If this semi-system
constitutes for us what is to be understood by "rationality"
in the sphere of conduct, I simply do not see how Rice
can demand more unity and order than it provides in the
name of "reason." If the philosopher cannot, in the face of
this, restrain his passion for making order, then let him make
it, but he will have passed beyond the bounds *\
at this point
of mere analysis or understanding, and, like Bentham, will /
have adopted the role of a legislator's legislator. He is free \ ^
to do so. But he cannot then reasonably complain if the rest \
of us do not choose to ride down the king's highway on his^
hobby-horse.
It will have been observed that I am not denying that
Rice's principle of intrinsic value is Nor do
a valid principle.
I contend that the principles of liberty and justice take
f
precedence over it. What I am objecting to is merely the

thesis that the dimension of "affective tone," as Rice some-! Y-
'
times calls it, is to be regarded as the only Identifying Prop- J
erty of intrinsic value. I s ay_that we live in an ethical plu ri-/

ve rse, not a univers e, andthat as long as weremai p as w ^
arj^Jjire's -Idenlifymg Prope rty will not be the only one to
which "inMnQir g^T" is the t rigger.
Rice's concluding chapter is called "Naturalism and the
Tragic Sense." No other contemporary philosopher, natural-
istic or otherwise, has better understood or faced more
unflinchingly the inescapable tragedy of human existence.
Like the great tragic poets, upon whom he discourses in
pages with so much sensitivity, he finds in tragedy

his last
both a metaphysical and an ethical dimension. The meta-
physical tragedy, you like, is the extinction of life; the
if
from man's incurable finitude and the

ethical tragedy arises
consequent waste of good. However, there is one aspect
of Sophoclean tragedy which, as itseems to me, Rice does
not adequately explain. In fact, I do not see how he could
explain it, given his doctrine of intrinsic value. The tragedy
of Antigone, in its ethical aspect, is not, at bottom, a tragedy
of conflicting maxims that might perhaps have been re-
moved, had Antigone or Creon been wiser, more selfless,
or more restrained. Here, as Hegel pointed out, is the
tragedy of spiritual self-division within the moral life which
cannot be removed by an appeal to some higher principle
of adjustment or harmony. In the case of Antigone, we are
not faced with a case of routine wasted good or sacrificed
virtue. Neither is it a case, such as we find sometimes in
Aeschylus or in Shakespeare, of the universe being out of
joint, so that man's deepest cry for justice, liberty, or mercy
goes unheeded by the natural or, if it exists, the super-

natural order. Nor, finally, is it a tragedy which can be
fully comprehended in terms of Aristotelian flaws. All of
these forms of tragedy are real enough, God knows. But
beyond them there is that special and supremely poignant
form in which the individual or the society is confronted with
an internal conflict of principle for which the moral order
provides no solution.
What this sort of is that while there is
tragedy teaches us
wisdom in the search for moral as well as intellectual order,
there is also a kind of wisdom and integrity that comes from
the recognition of the surds of the ethical life. The myth
of Genesis embodies a second great archetypal situation in
which Western culture has always been involved. Before the
Fall, man lived innocently, in joy and peace, but without
freedom and without dignity. In the one case, he was su-
premely happy, but lacked all sense of what it means to be a
person; in the other, he acquired personality and the
rights of conscience, and with them guilt and responsibility.
From the standpoint of the greatest happiness principle,

the price that he has paid for these human prerogatives
must seem exorbitant. From hard to see
this standpoint, it is
how the conclusion is be avoided that, if he had the
to
chance, man ought to return to Eden. But it is a conclusion
which dies on our lips. For we would not buy felicity, for
oursg lves or oth ers, if it required u"s~Eo" forego altog ether the
rights and responsibilities of free dom~and justice. In another
way, while we yearn for love, we demand respect, and if
love is offered in place of respect, it is a sinful love which
we must reluctantly oppose.
There are one or two other aspects of Rice's theory about
which I also have certain reservations, but I must deal with
them, now, very briefly. After several rereadings, I am still
not wholly satisfied by his account of the relations between
Conferring and Identifying Properties. In the first place, it
does not appear to me that the relation between them is, as /
Rice himself suggests, a relation of the less to the more y
general. The difference between the properties that make
things of a certain sort good in their kind and those which
have to do with right action is of another order. Rice does not
make as sharp a distinction as I could wish between those
virtues of things which answer merely to specific interests
such as gardening, fly-casting, and the selection of judges,
and the standards which, even provisionally, prescribe what
we think of as moral duties. There is, in brief, a functional
difference between, on the one hand, forms of grading
which establish the criteria for evaluating or commending
things that answer to particular wants and needs and, on the \
other hand, the maxims which we call "moral," the role of )
which is to prescribe, in our dealings with other persons, our ( is
special duties or obligations to them. In justifying the latter, J

we do undoubtedly appeal to higher level ethical principles
of intrinsic value. But it is not clear to me that such princi-
ples are relevant when questions are raised about the stand-
ards that are used for grading dry-flies or garden-hoses.
It is only when we have to decide whether we ought to cast
flies or to garden that questions of moral principle and hence
of intrinsic value need arise. The grading of tools and the

appraisal of specific "goods" are normally tied to particular
/\activities and concerns which, as such, do not raise questions
of conduct. It is for this reason, so far as I can see, that it
/ is necessary to distinguish between ethics proper and the

\ general theory of valuation. Certainly, Rice himself has rec-
ognized this distinction in his previous writings. I am in-
clined to think that he, like Perry, has drawn it in the wrong
way.
[XIII]
History, Morals,
and the Open Society
In the year 1909, Thorstein Veblen summed up the method

which at long last would transform the "sciences of man"
from essays in ideal social mechanics or mere empirical
agglomerations of "facts" into genuine sciences capable of
arriving at laws of social behavior and development.
In so far as modern science inquires into the phenomena of life,
whether inanimate, brute, or human, occupied about ques-

it is
tions of genesis and cumulative change, and it converges upon

a theoretical formulation in the shape of a life-history drawn in
causal terms. In so far as it is a science in the current sense of
the term, any science, such as economics, which has to do with
human conduct, becomes a genetic inquiry into the human
scheme of life; and where, as in economics, the subject of in-
quiry is the conduct of man in his dealings with the material
means of life, the science is necessarily an inquiry into the life-
history of material civilization, on a more or less extended or
restricted plan. . . . Like all human culture this material civ-
ilization is a scheme of institutions — institutional fabric and in-
stitutional growth. 1
—
Here are the key concepts "genesis," "cumulative change,"
'life-history drawn in causal terms" —
the application of which,
1 "The Limitations of Marginal Utility," 1909. In What Veblen

Taught, edited by W. C. Mitchell (New York: Viking Press, 1945),
pp. 162-3.
in the view of Veblen and his contemporaries, was to raise

inquiries into human behavior from the "taxonomic ' stage to
that of "evolutionary science." Yet not four decades later,
thissame method, under the epithet "historicism," has be-
come the flogging horse of advanced critics who regard it as
a prime reason for the failure of social theory to achieve the
status of genuine science. With the appearance of Professor
Karl Popper's book, The Open Society and Its Enemies,2 this
revolt against "the historical method" enters into the stage
of a full-blown war against all attempts to understand man
and his institutions in terms of the "laws" of their historical
development. But the enemy no longer limited to latter-
is
day historicists. Nearly the whole first volume of Popper's book

is devoted to tracing in Plato's dialogues —
and especially in
—
the Republic the reactionary moral and social theories in
which germinated the central ideas of an "evolutionary
science" of political and social change.
Unlike Professor John Wild, who has found in Plato's
Republic the perfect anagram of the "good" that is, the —
—
democratic, classless society, Popper discovers there the per-
fect prototype of tribalistic collectivism. For Wild, Plato was
the first to discern the true, rational, and natural order of
things, the inversion of which leads to all the ills, spiritual
and material, that flesh is heir to. For Popper, Plato's "Reason"
is a reversion to the methods of "insight" and "prophecy"
which are characteristic of pre-rational mythological modes of

thought. Yet, ironically, both Wild and Popper, in their
respective defense of and attack on Plato, regard themselves
as defenders of the faith in free and rational society.
This antithetical partisanship reaches its climax in their
contradictory interpretations of the "ideal" of the philosopher
king. For Wild, this conception is but an idealization of that
rule of reasonand virtue which is the aim of all true democ-
racy. For Popper, the doctrine of the philosopher king is a
not too subtly disguised bit of propaganda on Plato's part
for his own claim to kingly power in Athens.
In a polemical age it is perhaps not permissible simply to
2 4th edition (London: Routledge and Kegan Paul, Ltd., 1962).
History, Morals, and the Open Society 265
sit on the sidelines and cheer. Let me say, then, albeit reluc-
tantly, that Popper's caricature seems to me somewhat more
recognizable than Wild's. For my part, however, I should
prefer to remember the great dualist who gives us "the free
interplay of ideas," each of which embodies a part, but only
a part, of the truth. It is this Socratic Plato, for whom the
polar ideals of order and freedom, unity and difference,
personal integrity and collective security, individual con-
science and public opinion, quality and quantity, all have
their inviolable rights in any adequate social order. This
Plato, I believe, would have seen Popper's antithesis of
"open" and "closed" societies represent not simply a conflict
of light with darkness, of good with evil, of tribalism with
human dignity, but the complementary sides of one economy
of values, which must include them both as counterpoises.
The liberal mind, in my opinion, has nothing to do with
such inordinate and partisan zeal as Popper so frequently
displays in his book. His method can only stultify the under-
standing of our intellectual heritage. Its aim is not merely to
criticize and correct what are considered to be the lawful
errors of one's fallible predecessors, but also to discredit
them in toto by inference and innuendo. They are treated,
not as "the loyal opposition," but, in many instances, as
men ridden with ambition, as sycophants and mystagogues.
Their doctrines are construed not as honest attempts at
evaluation and understanding, but as deliberate obscurant-
ism and intellectual chicanery. This is true of Popper's treat-
ment of Plato and Aristotle. It reaches its climax in his
treatment of Hegel, whom
he regards, following Schopen-
hauer, not merely as an intellectual clown, but as a paid
mercenary of the King of Prussia. Only Marx, of the major
figures discussed, is recognizable as something other f:han an
"enemy" in the literal sense. Marx was, Popper concedes, an
honest man, a genuine scientist, and a true humanitarian. He
was simply misled, by his addiction to "historicism," into
development and
unscientific formulations of laws of social
ungrounded prophecies concerning the future of democracy
and capitalism.
ECONOMIC PLANNING AND

FREEDOM OF CONSUMPTION
In the remainder of this essay I will concern myself no fur-
ther with Popper's interpretations of the great historical
figures mentioned above, since it is not with them but with the
ideas they represent that the more interesting portions of
Popper's book are concerned. I will discuss, instead, several
of Popper's own theses which seem to me of great importance

to contemporary social philosophy.
At the outset, it is well to call attention to an openly
avowed relationship which is important for a correct under-
standing of Popper's whole point of view in social philosophy.
This is his connection with Friedrich von Hayek, many of
whose ideas Popper
find a close parallel in Popper's work.
acknowledges his indebtedness to Hayek in the "Acknowl-
—
edgments," "without his interest and support the book
would not have been published." It is not hard to see, I think,
why Popper should have found Hayek such a determined
supporter. Popper's objections to what he calls "Utopian en-
gineering" are even more extensive than Hayek's criticisms
of centralized or collectivist planning. Popper refers approv-
ingly to Hayek's criticisms of such forms of planning on the
ground that they "eliminate from economic life some of the
most important functions of the individual, namely, his func-
tion as a chooser of the product, as a free consumer." 3 Pop-
per's own criticisms in part also follow this line.
Nowhere, however, is there any sustained analysis of the
alleged "technological impossibility" of a planned system of
production in which there also would be "freedom of con-
sumption." Popper ignores the point which many economic
collectivists regard as essential, namely, that without organ-
ized and planned distribution of commodities the masses,
in such countries as India and China, are often powerless to
procure even the basic necessities of life. He also forgets that
even in a "free economy" effective choice is often limited
3 Popper, op. cit., Vol. I, p. 242.
to a selection of alternative labels rather than different types

of product. Popper comments on money as a "symbol" of
the open society. 4 Money, according to him, is "part of the
institution of the [partially] free market, which gives the
consumer some measure of control over production." 5 But
he does not seriously consider the means by which a "free"
economic system can secure a more equitable distribution
of money so that the people may effectively and continually
exercise the freedom of choice about which he is so much
concerned. It is true that the Marxists fail to realize "the
danger inherent in a policy of increasing the power of the
state." 6 But Popper too easily passes from the idea of collec-
tive planning to that of a totalitarian or closed society. More-
over, when it suits his purpose he is quite capable of ar-
guing, as he does here, in terms of the "essentials" and "in-
herent characteristics" which he finds so objectionable when
employed by his opponents, whether Aristotelian or Canta-
brigian.
THE CRITIQUE OF HISTORICISM

The central concept of the book is something called "his-
toricism." Indeed, it is the exploration of the many-sided
consequences of this idea which provides the book with
whatever unity it possesses; and Popper's most serious objec-
tions to the major historical figures discussed 7 are based on
their adherence to "historicist" principles.
The critique of historicism actually proceeds on three
main levels: (a) an analysis of the effects of historicist
assumptions upon the social philosophies of Plato, Hegel, and
Marx; (b) an analysis of the usefulness of the historicist
method as a means of deriving laws of historical develop-
ment from the observation of social phenomena; (c) a dis-
cussion of the implications of historicism with respect to the
4 Ibid., Vol. I, p. 266.

« Ibid.
« Ibid., Vol. II, p. 121.
7 That is, Plato, Hegel, and Marx.
possibility of social planningand control. According to Pop-

per, if we
accepted the fatalism implicit in the historicist
theory of laws of social development, we would logically be
driven to deny the possibility of human choice and hence
of the piecemeal planning and control of social forces in
which, according to him, lies the salvation of mankind.
Historicism appears in its earlier (as well as some later)
forms in the guise of the myth of the chosen people, selected
by fate to inherit the earth, or to lead mankind to the class-
less Utopia. Later it is transformed into the "scientific doc-
trine" that laws of historical development can be determined
which will enable us to predict the future course of social
change. Popper rejects both theistic fatalism and the later
"scientific" historicaldeterminism. Sweeping historical proph-
ecies, he maintains, "are entirely beyond the scope of
scientific method." 8 "The future," he insists over and over
again, "depends on ourselves, and we do not depend on any
historical necessity." 9 What Popper advocates is a type of
social engineering which asks no questions about historical
tendencies and human destiny. He believes that "man is the
master of his own destiny, and that in accordance with our
aims, we can influence or change the history of man just as
we have changed the face of the earth." 1
Here we may perhaps discern a certain ambivalence in
Popper's position: We are the masters of our destiny, but only
so long as we radically limit our conception of what that
destiny is to be; we can predict and control in a limited way
what will happen next, but we can not predict or control
long-range social developments. In this way, as it turns out,
there is a sense in which the historicist "utopian engineer,"
believing as he does in the possibility of long-range predic-
tions and the desirability of long-range blueprints for dis-
tant objectives, has a more complete faith in man's mastery
of his destiny than Popper has.
In what, precisely, does the basic error of historicism
8 Popper, op. cit., Vol. I, p. 3.

9 Ibid.
1 Op. Vol.
cit., I, p. 17.
consist? In order to make this clear, it is necessary to refer

briefly to Popper's conception of causal explanation and
prediction in the "generalizing sciences such as physics and
biology." Causal explanation involves the framing of gen-
eral hypotheses (universal laws) which, together with cer-
tain specific (or singular) statements of the initial conditions
which pertain to the special circumstances of a given case,
enable us to deduce a statement (prognosis) of what will
happen to the case in question. The initial conditions are
regarded as the "cause" of the event, and the deduced re-
sults are the "effect." Thus we can only speak of an event
as a "cause" relatively to some universal law; when the pre-
dicted event is observed actually to occur, the theory itself
is then said to be tested or confirmed. Now the pure general-
izing sciences are not interested in the prognosis as such.
They are concerned with it only as a test of some law or
principle. In the case of applied science, however, the interest
isin the prognosis itself. Hence the difference between "pure"
and "applied" generalizing sciences is primarily a difference
in interest. "Whether we use a theory for the purpose of
explanation ... or of testing, depends on our interest, and
on what propositions we take as given or assumed." 2 It
follows from this that applied generalizing sciences are con-
cerned to predict specific or particular events. 3 It would seem
to follow also that the interest of such sciences is, in part at
least, historical, since according to Popper the so-called
historical sciences are interested "in specific events and in
4
their explanation." Nonetheless, and this is the puzzling
point, Popper tells us that "the sciences which have this in-
terest may, in contradistinction to the generalizing sci-
. . .
ences, be called the historical sciences." 5 But how is this

statement to be reconciled with his preceding remarks about
applied generalizing sciences? To me, at least, the answer
remains obscure.
2 Op. Vol.
cit., II, p. 250.
s Ibid.
4 Op. Vol.
cit., II, p. 251.
5 Ibid.
He goes on, at any rate, to say that "from our point of view,
6
there can be no historical laws." Generalization simply be-
longs to a different line of interest. This seems to be the sole
basis of Popper's criticisms of Professor M. G. White, whom
he "what has been described here in
criticizes for neglecting
the text as the distinction between historical and generaliz-
ing sciences, and their specific problems and methods." 7
But such an exclusion, purely by definition, of "historical
laws" is simply verbal. It sheds no light whatever on the
employment of the historical method or the possibility of
laws of development which apply to historical processes. It
is also misleading, since it has nothing to do with other, more
serious difficulties which Popper raises with respect to the use

and prediction.
of history for the purposes of generalization
However, the two are constantly run together and confused
in Popper's exposition.
In order, therefore, to distinguish between Popper's real
and merely verbal objections to historical laws, we must
his
."distinguish, asPopper unfortunately does not in any clear
f
way, three different questions: (a) whether historical data
can be trusted for evidence for social laws, historical or other-
wise; (b) whether there are verifiable laws of development
upon the basis of which we may predict future events ( and
confirm our laws); and (c) whether there can be what may
be called laws of "unrestrictive scope" in terms of which all
social processes may be explained. The denial of c does
not entail the denial of b nor of a, nor does it render b
trivial, as Popper suggests. On the other hand, the denial of
a would, as we shall see, render a positive answer to both
b and c impossible. But Popper's emphasis on the uniqueness,
unrepeatability, and uncontrollability of historical data sug-
gests, at the very least, that they cannot be trusted, and
hence, as we shall see, that there can be no trustworthy
social theories whatever. This despite the fact that he
raises no doubts at all about the possibility of sociological,
6
Op. cit., Vol. II, p. 251.
Op. cit., Vol. II, p. 344. Cf. M. G. White, "Historical Explanation,"
7
Mind, Vol. 52, 1943, pp. 212 ff.
economic, or political laws, but only about laws of historical

development.
(a) To answer the first question, it should first be pointed
out that Popper nowhere denies that social theories can be
successfully used by the historian in validating records and
in ascertaining the reliability of historical data. Granted that
the margin of error is always very considerable, it would seem
that if they can be used at all, they can be used for the
confirmation of "historical" or any other type of social theory.

The trouble with such data is of three sorts: they can not
be varied or repeated at will; they are always a very limited
selection of all possible relevant data and are based upon the
"preconceived" interests or theories of those who select
them; and since no further facts are available, the testing of
8
theories "will not, as a rule, be possible."
These are serious charges, and I am sure most historians
and social scientists would agree that they present very im-
portant obstacles to the verification of social hypothesis. But
are they any more serious for historical theories than for
merely social theories in economics, sociology, or politics?
If "the so-called 'sources' of 'history only record such facts
as appeared sufficiently interesting to record, so that the
sources will on the whole contain only facts that fit in
with a preconceived theory," and if "since no further facts are
available, it will not, as a rule, be possible to test that or any
other subsequent theory," 9 are not the consequences of this
quite as devastating for one type of social hypothesis as for
another? The charge of circularity in the case of historical
theories would, if we agree with Popper, apply also in the
case of sociological and economic theories.
According to Popper the facts in science are always col-
lected with an eye upon the theory, but they confirm the
theory "only if they are the results of unsuccessful attempts
to overthrow its predictions, and therefore a telling testi-
mony in its favor." 1
It is, he insists, "the possibility of over-
8 Popper, op. cit., Vol. II, p. 252.

9 Ibid.
1 Op. Vol.
cit., II, p. 247.
throwing it, or its falsifiability, that constitutes the possibility

of testing and therefore the scientific character of a
it,
2
theory." Popper does not deny that such possibilities are in
fact open to the social scientist, despite his disparagement of
the historical source material upon which all social sciences
depend. Yet he regards this as a conclusive argument against
historical laws.
(b) The second question, whether there can be laws of
development, must, I think, be answered in the affirmative
by anyone who admits that there are dynamic laws of
society, i.e., laws of succession rather than coexistence.
Whether such laws are called "historical" or "sociological" is a
question of terminology. If such laws are, as they must be,
largely based on historical data, and if they are used for the
explanation or prognosis of other historical events, future
or otherwise, then seems not inappropriate to speak of
it
them as historical as well as sociological. Popper himself

explicitly admits that there can be sociological laws, even ones
pertaining to the problem of progress. 3 But if this is so, then
to argue that "we should better not speak of historical laws
at all" merely on the ground that when we are confined to
one unique process there can be no law of nature, or that
history is interested in specific events rather than general
laws, 4 is to quibble (inconsistently, as we have seen) about
the use of a word. Why should it be assumed that all his-
torical events are unique (i.e., absolutely different from all
other events ) or that the interest in generalizations and their
employment in explaining historical events is beyond the
scope of the historian? Such assumptions are clearly in
flagrant inconsistency with what the historians have found
and done. If Popper argues that it is not within the scope of
the historian, as historian, to concern himself with such
matters, then I reply that he himself is taking the "essential-
ist" position which he has so brilliantly criticized elsewhere
in his book.
2 Ibid.
3 Op. cit., Vol. II, p. 306.
* Ibid.
(c) The third question, whether there can be a "universal

point of view for history," can easily be answered in the
negative. But it does not follow from this that the laws of
society, historical or sociological, are not quite general.
Popper seems to me to confuse the generality or universality
of a proposition with its scope. For example, any proposition
is general if it asserts that all of the members of one class are
members of a second: "All men are mortal" is quite general,
even though there are many things to which it does not apply.
Such a proposition is limited in its scope in the sense that
there are many things not explained by it simply because
they do not belong to the class of men.
The "infinite subject-matter of history" doubtless renders it
impossible to obtain hypotheses which are not of limited

scope. Therefore, as Popper says, we cannot avoid a limited
point of view in history; we must, that is to say, limit ourselves
to "economic" laws of history, "political" laws of history,
Popper himself is very partial
"military" laws of history, etc. (
which assert a universal

to political laws, especially those
tendency to the abuse of power.) To this most of us would
undoubtedly agree. Those writers such as Hegel, Marx,
Spengler, and perhaps Toynbee, who have attempted to
formulate general social laws of unrestricted scope, in terms
of which every historical event is to be explained, are de-
serving of Popper's strictures against them. But not all his-
toricistshave attempted anything so grotesquely speculative.

Nor does the failure to achieve a universal point of view for
the universe itself impugn the importance of laws of histor-
ical change, any more than the failure to achieve a universal
point of view for the universe itself impugns the importance
of the laws of physics. Triviality is not necessarily commen-
surate with limited scope.
We may then agree with Popper that Mill was probably
wrong in supposing that we can "find the law according
to which any state of society produces the state which suc-
ceeds it, and takes its place." 5 But the implications of this
5 Quoted by Popper, op. cit., Vol. II, p. 83.

are surely as momentous —and —for

as trivial one science
of society as for another. They are surely no more so for
history than for economics.
It would seem, then, that Popper's wholesale indictment
of "historicism" begins, after a while, to wear the aspect of
an intellectual witch-hunt, rather than a merely objective
criticism of presumptuous hypothesizing and ungrounded
large-scale prognostications about what "must" happen in the
future.
To return, now, to Popper's contrast between Utopian and
piecemeal engineering, we may agree that large-scale plan-
ning for distant objectives is likely to be both unscientific and
difficult to implement. It is safer and easier always to plan on
a limited scale. But without long-range objectives, the hope
for organized and ordered social change must give place
to ad hoc remedies for specific immediate ills. Whether such
remedies are desirable, however, in the light of more general
social purposes, would, if Popper is right, be impossible to
determine. Actually the effect of Popper's analysis is to dis-
credit the practical use of our rational faculties when ap-
plied to anything but the simplest and most immediate of
results. Its social consequence, I fear, is a deep conservatism
and an opposition on principle to bold policy-making of any
sort in the interest of the people as a whole.
The fact is, however, that we cannot escape making large-
scale predictions and plans. The reasonable procedure
seems, then, not to renounce any attempt at prognosis con-
cerning the future development of society, but to try to
deduce the margin of error in such prognoses as much as
possible by constantly improving our interpretations of his-
tory. By denying them the title "scientific" we might incul-
cate a proper humility and scepticism with respect to our
guesses. But we must also remember that there is an enor-
mous difference between crystal-ball gazing and responsible
interpretation. In the end the task of the critic of social
theory is not to persuade men to forswear long-range plan-
ning, but to make them conscious of its extreme difficulties,
and the urgent necessity of keeping plans flexible enough to
accommodate the unanticipated changes which they may

have to face.
mSTORICISM AND ETHICS

There are certain other aspects of Popper's attack on his-
toricism, however, with which I am in somewhat closer agree-
ment, especially as it applies to the problems of ethics, al-
though even here Popper strikes me as being at once doc-
trinaire and misleading. Now, according to the historicist point
of view, moral categories are wholly relative to particular
historical situations. Is it "right" to keep one's promises? To
this question one can only reply: Yes and no; yes, if you are
speaking from the standpoint of a certain particular society
or period; but no, if you generalize with respect to man-
kind in general. Any moral system applies only within a cul-
tural-historical epoch, society, or class.
The interesting point about Popper's rejection of moral rel-
ativism, however, is that it goes with an insistent rejection
of (a) the possibility of a scientific ethics, and (b) the
verifiability of normative statements of any sort. The
grounds of his rejection of a system of ethical norms con-
structed on a scientific basis are of considerable interest. In
the first place, "if it could be achieved, it would destroy all
personal responsibility and therefore 6
all ethics," this despite
the fact that science has nothing to do with determinism. Sec-
ondly, only "a scandalmonger" would be interested in such
judgments as applied to other people; "'judge not' appears
to some of us one of the fundamental and much too little
7
appreciated laws of humanitarian ethics." Thirdly, the at-
tempt to base an ethics on "human nature" leads nowhere,
since all founded upon human nature, so that the
actions are
moral problem invariably becomes the question: "which ele-
ments in human nature I ought to follow and to develop, ,/
and which sides I ought to suppress or to control." 8 Fourthly, .
6 Popper, op. Vol.

cit., I, p. 207.
7 Ibid.
8 Ibid.
zy6 Reason and Conduct
\any analysis of the word "good" would fail to answer the

£ question: "Why ought I to concern myself with that?"
He concludes that all discussion about definitions of good
are useless. Indeed, even the attempt to determine stable
cognitive meanings, and hence verifiable ethical judgments,
isan escape from the realities of the moral life; i.e., our
moral responsibilities. But, of course, the question remains:
"What are our moral responsibilities?" Is it possible to "know"
them? And if I elect to ignore them ("Why should this con-
cern me?" ), then what? It is all very well for Popper to
deny the possibility of objective ethical judgments, but then
he must, it seems to me, eschew talk about moral escapism,
evasions of responsibilities, and the evils of the closed
society. Is he simply blowing off steam, or trying to goad us
into agreement with him? After all, why should we agree?
In his view, is it even possible to make sense of this ques-
tion?
Popper devotes a whole chapter (Chapter V) to a dis-
cussion of the distinction between natural laws and norma-
tive laws or rules of conduct. The distinction is for him
"fundamental." He contrasts the "naive monism" of the pro-
ponents of the closed society with the "critical dualism" of
the proponents of the open society. The former makes no
distinction between natural and moral law. The latter, which
accompanies the breakdown of magic tribalism, sees that
human laws are "made by men." Norms are "man-made" in
the sense that we "must" (sic) blame nobody but ourselves
for them. They are not necessarily "arbitrary"; they can be
improved; some are "better than others." 9 But in saying this
we are in no sense stating a fact.
Now Popper is obviously right in insisting upon the im-
portance of the distinction between natural laws and man-
made laws. In fact it is nearly incredible that anyone should
have been confused on this point merely because of the his-
torical accident that the word 'law" is employed ambiguously
in referring both to well-established scientific theories and
to the positive laws of politically organized societies. But it
9 Op. cit, pp. 51-2.
is no anyone should be confused by the

less astonishing that
term is also employed ambiguously
historical accident that the
in speaking both of positive laws and of the principles or pre-
cepts that serve as guides to personal moral judgments and
decisions. What certain philosophers, ecclesiastical digni-
taries, and holiday orators call "the moral law which stands
above the laws of men" is certainly not a law of nature, and
the whole "natural law" tradition in moral and legal theory
was exploded the moment Hume unpacked the logical con-
fusion on which it rests. But it does not follow from this that
moral principles are to be classified, along with positive laws,
as man-made. Nor is this confusion mitigated by the fact that
neither moral principles nor positive laws are falsifiable or
testable by the procedures of empirical science. A great many
forms of utterance, including most philosophical theses,
works of literary art, many and
interpretations of history,
religious creeds are not testable by such procedures. But
what else do they share in common? Here, as it seems to me,
we are faced with the age-old tendency of the rationalistic
lump together indiscriminately any and all propo-
tradition to
sitionswhich happen not to convey testable information
about something which is called, in some privileged sense,
"reality." 1
The reason for this, I am convinced, is not that
the representatives of this tradition are in possession of
sounder, logically more coherent theories of meaning and
knowledge which enables them to strip these other forms of
discourse of their false or misleading pretensions, but rather
any proposi-
that the rationalists are not really interested in
tion which does not purport knowledge of
to increase our
"reality." They are, in effect, "gnostics" who consider any
form of theoretical activity unrespectable for which no "cog-
nitive" claims can be made. And it is for this reason that
they are so obsessed with what Popper elsewhere calls "the
problems of demarcation": the problem of finding a general
criterion by means of which we can distinguish between
propositions which convey knowledge of reality and those
1 The great question here, from a logical point of view, is whether

they are entitled to their privileges.
which do not. Any other problems of demarcation simply

do not interest them. In this respect, at least, Plato and
Popper are identical twins! Popper calls it the problem of
demarcation; Plato calls it the problem of the "divided line."
Once this problem has been solved to their satisfaction, their
interest in analyzing what lies on the other side of the divided
line simply disappears. And they are both content with
catch-as-catch-can remarks about the "non-cognitive" dis-
course of poets, politicians, lawyers, divines, and moralists.
Plato regards them, in effect, as liars and pretenders. Popper
is more discreet. But the difference between them is less
logical than rhetorical.
What Popper fails to see is that the very "critical dualism"
of which he makes so much is itself largely responsible for the
tendency of many who reject the notion that moral prin-
ciples are "man-made" moral principles to the
to reassimilate
"laws of nature." For example, Plato, whom Popper classi-
fies as a "monist," understood quite as well as Popper the
difference between man-made laws and the laws of natural

science. It was indeed this very difference which formed the
basis of his harsh criticisms of the Sophists, for whom Popper
mas such a tender regard. Plato correctly perceived the error
/involved in classifying moral principles, as the Sophists had
Vdone, as forms of man-made law, but as a critical dualist
he was forced to conclude that they must therefore be a kind
of natural law which conveys information about some trans-
empirical realm of being. His error is precisely a mirror-
image of Popper's own. As a critical dualist, Popper's prob-
lem is no different from Plato's, but perceiving the equally
profound error of classifying moral principles as laws of
nature, he, like the Sophists, is obliged to regard them as a
kind of man-made law. And because he so treats them, he,
like the Sophists,makes it appear that his own libertarian
moral convictions are no less arbitrary and no less conven-
tional than those professed by the proponents of the closed
society. At the same time he seems to undercut any clear,
logical basis for a moral critique of positive law itself,
thereby disposing his critics to think that the difference be-
tween the closed society and the open society is merely a

difference in regard to the conventional forms of justice
which in each society determine what is lawful.
Here it strikes me that the intuitionists, whom Popper so
much abhors, show a nicer sense of discrimination. What-
ever their other faults, their fidelity to the familiar forms of
moral discourse saves them from the logical confusions which
Popper's own legalism and decisionism involve. They ob-
committing a solecism we
serve, quite properly, that without
cannot speak of "making" our moral principles any more than
we can speak of making the laws or nature. They point out
also that we do not ordinarily speak of "deciding" what
principles we ought to live by, but rather of believing or
knowing that we should live by them. Popper's analysis of
morals makes it appear to be a sheer verbal anomaly that,
whereas it makes no sense to ask whether a positive law, a
social convention, or indeed any "man-made" rule which
rests upon a decision, is true or false, it is considered proper,
in the moral sphere, to speak of true or false judgments, of
knowing what is right and good, and of discovering certain
moral truths which one did not know before. The intuition-
ists, on the other hand, accept these linguistic facts as in-
dicative of interesting and important logical characteristics of

moral discourse which any adequate moral theory must at
least try to explain. In season and out, they insist upon the
meaningfulness of our ordinary questions concerning the
objectivity of moral judgments and the validity of the justi-
fications we offer in their defense. Not only does Popper offer
no explanation of these features of moral discourse, his
radical non-cognitivism and subjectivism forces him system-
atically to discount them in effect as so many evidences that
ordinary moral thought, as reflected in our prevailing lan-
guage of morals, is nothing more than the hangover of a
tribalisticverbal magic which clear-headed scientific philoso-
phers will have nothing to do with. They are, of course,
entirely free to do so, just as they are also free to renounce
theology or any form of discourse other than those of em-
pirical science and mathematics. But then they can hardly
expect to escape rebuke from the theologians or moralists

when they take a holiday from their scientific labors in
order to play at linguistic games whose rules they despise.
These remarks are by no means intended to imply that I
agree with intuitionistic explanations of the meanings of
ethical terms and judgments. Far from it. As I have ex-
plained elsewhere, the intuitionists are themselves radically
misled by gratuitous assumptions concerning the nature of
meaning, truth, and definition into the construction of hypo-
and epistemological mythologies which I, like
static ontologies
Popper, want no part of. However, I want no part of them
because they are, in my opinion, completely unnecessary.
What I do contend is that, at what may be called the phe-
nomenological level, they describe the forms of moral
thought and judgment far more truly and accurately than
Popper. Popper's own uncritical version of critical dualism
precluded him from making good sense of the very forms of
words which, in practice, enable ordinary men to preserve a
clear and radical distinction between moral principles and
man-made laws or decisions, between questions of moral
justification and those which are addressed to matters of
conventional propriety and good Let me say also that
taste.
no more than Popper do I regard ordinary language as a holy
of holies. I am saying only that before its rules are amended
they should be correctly understood, and that Popper, like
most other scientific philosophers of our age, has simply not
taken the trouble to study them with any care. Because of
this, he casually and uncritically adopts the makeshift di-
chotomies of "critical dualism" which prevent him not only

from clearly articulating his philosophical disagreements
with Plato and other proponents of the closed society but also
from making clear to his readers what are the fundamental
moral bases of his own defense of the open society. Thus he
makes it appear, tragically, that the man-made morality of
the open society is just as arbitrary, just as willful, just as
subjective, and, at bottom, just as much a matter of individual
taste as that of the totalitarians.
By a strange irony, Popper's moral theory thus leads him
to the verge of another mistake which is also analogous to

one which he ascribes to Plato. In pointing out this error,
I should like to emphasize that I am merely following Popper's
own practice of tracing undesirable moral stances to philo-

sophical theories about morals which, apparently, belong to
a completely different level of discourse. I do not object
one does not confuse the intention
to this practice, so long as
with the effect of a theory and so long as one bears con-
stantly in mind that in judging the "consequences" of the
theory one should not uncritically beg the very point at issue.
In this sphere it is not easy to say where questions of
"logic" leave off and questions of "ethics" begin; that, indeed,
is just the point. Popper's error, like Plato's, is "global,"
depending as it does upon a complex of logico-ethical beliefs
which are hard to disentangle from one another and whose
"formal" implications, therefore, are next to impossible to
determine.
Now it seems to me Popper shows great insight in
that
pointing out the moral and political aestheticism which
stems from Plato's reduction of all moral and social ques-
tons to questions about what appears most harmonious, "just,"
and "fitting" to a detached, impersonal observer who, for-
tunately for himself, doesn't have to live in the picture whose
elegance and unity he so greatly admires. This aestheticism,
as Popper shows, is closely related to the views that moral
and political issues are essentially problems of knowledge,
that in knowing something one's love of it is consummated,
and that knowing anything, including the good, is essentially
a matter of impersonal contemplation or intuition of its
ideal form. To adopt a convenient phrase of Professor Gil-

bert Ryle, Plato is thus involved in a series of monumental
"category mistakes" which lead him ineluctably to judge
social systems, not from the moral standpoint of human
rights and sufferings, but from the aesthetico-cognitive point
of view of harmony, coherence, and clarity. Thus also Plato
is led to construct his ideal republic as if he were com-
posing a picture or proving a theorem, rather than helping

his fellows to find a way out of the land of Egypt, out of the
house of bondage. And it is for this reason that the Hebrew

prophets are in the end morally and politically so much wiser
and so much more moral than he.
Here, it must be understood, we have to do with questions
not only of substantive precept but of underlying method,
not only of judgment but of the whole philosophy which
animates and guides that judgment. So it is in Popper's case
as well. Popper will have nothing to do with the "logical"
thesis that morality is a form of natural law. Strictly speaking,
moral principles are not falsifiable, and, accordingly, are
neither true nor false. Moral laws belong to the class of
man-made laws, and in making a law the fundamental prob-
lem is to determine what one really likes. But this surely
— —
suggests if indeed it does not imply that every funda-
mental question of ethics is a question of taste. And, as
everyone knows, about tastes there is simply no disputing.
Popper's own moral taste, as it turns out, is not for orderly,
well-proportioned social structures in which there is a place
for every part and every part keeps its proper place. Rather
does it resemble that of a romantic artist who, disdaining
contemplation, is interested only in doing and making some-
thing which expresses his own heart's desire. If it happens
not to look well to a disinterested observer or critic, why so
much the worse for him. What has the art of morals to do
with agreeable proportions and harmonies? "True" morality
is not a matter of judgment but of action, not a matter of
knowledge but of decision, not a 'matter of perception

and understanding but of feeling.
There is a further aspect of Popper's conception of morality
which remains to be mentioned. It is not unrelated to his
attack upon historicism, I think. Now it is characteristic of
many historicists, such as Hegel, to emphasize the extent to
which the consciousness of individual persons reflect and are
determined by the historical institutions of the societies to
which they belong. Accordingly, they tend to view morality
not so much as a matter of the independent, critical judgments
of individual men who have managed to break out of the
social womb in which they have been nurtured, but rather as
a code of traditional precepts which tie the individual umbili-

cally to forms of activity whose authority is completely im-
personal and customary. For them, morality tends to be
viewed as part of the "constituton" of an "objective" social
system, the criticism of which amounts in effect to a revolu-
tionary rejection of the system itself. It thus becomes in-
creasingly difficult for them to distinguish between moral
criticism and legal outlawry or between righteousness and
conformity to the requiements of the traditional code of
society. Popper, on the other hand, begins, a-historically, with
an idea of morality as a matter of man-made laws and
decisions. And he concludes accordingly that what man hath
made he can also unmake, and what others have decided,
every individual is Hence his opposi-
free to decide against.
tion to historical determinism. From such a standpoint, the
historicist is made
to appear as the advocate of a closed
society and opponent as the advocate of the open society.
his
Of course, there is no purely logical reason why an advo-
cate of Popper's ethical theory should not decide for himself
that conformity is always the best policy and that slavish
adherence to the traditional constitution is the only right

thing to do; indeed, so far as the sheer logic of the matter
is concerned, there is no reason why a critical dualist like
Popper should not decide for a completely closed society

within which no person, including himself, would henceforth
know the difference between conformity and conscience or
between criticism and revolution. On the other hand, there
is no reason in logic why an historicist, if he can extricate
himself from the influence of traditional ways, should not

conclude in conscience that they are evil ways, and hence
that the moral constitution of the community ought to be
abrogated. The historicist is also free to observe that in some
cultures "openness" is itself traditional, and that in upholding
the sacredness of the constitution of an open society one
may help to make it certain that self-appointed lawmakers
will not decide to introduce new laws which will henceforth
make it illegal to think and speak as a free man.
My point is this: Just as proponents of the doctrine of
natural law may be found at any given time in the camps

of reaction, liberality, or revolution, both proponents of
historicism and proponents of Popper's critical dualism may
be found there also. The real differences between them, so
far as morality, politics, and law are concerned, are on a
different level. The proponent of natural law, ignoring the
historical differences among men and institutions, is likely
to talk formalistically of certain abstract natural rights of
men without regard to their changing needs and individual
capacities. Thus, he is likely to be unwise and arbitrary in
his defense of principles which in one certain historical situa-
tion are morally necessary but in another are morally im-
possible. The doctrine of natural law tends to make men
dogmatic, and peremptory in their moral judg-
inflexible,
ments and deliberation. Analogous weaknesses are implicit in
Popper's own position. Abstract natural rights and a-historical
man-made decisions are in fact two sides oi the same coin.
Both ignore the existential context in which every moralist
finds himself and which no moralist may with impunity ig-
nore. The proponent of natural law declines to make a truce
with contingency; Popper, in the name of freedom, refuses to
make a truce with necessity. Both are blind. The one is the
philosophical progenitor of the paper-constitutions which
loudly but ineffectually proclaim "the rights of man," the
other of libertarian platforms and statutes which are incapa-
ble of implementation or enforcement. By comparison, it
seems to me, there is something to be said for the historicists.
The historicists' dull "grey on grey" is not inspiring: some-
times it and indiscriminate blurring
leads to the indecisive
of moral distinctions which Popper rightly deplores. But it
also saves them from the flat moralistic and legalistic blacks
and whites which obscure the practical problem of relating
discussions of moral rights and responsibilities to the concrete
historical conditions which determine whether any right may
realistically be claimed or any responsibility may sensibly
be assumed.
But Popper is not all of one piece, any more than is Plato
or Hegel. In other ways one finds strange parallels between
some of his own substantive moral views and those of the very
historicists whom he excoriates. For example, one of the fun-
damental lessons of Hegel is the futility as well as the pro-
vinciality of most abstract retrospective moral criticism
which spends itself in brooding judgment of crimes com-
mitted against the abstract name of "justice," rather than in
responsible deliberations which have in view the removal of
existing inequalities and injustices acknowledged by the
constitution of the community within which one lives. This
corresponds quite closely to Popper's own insistent claim that
moral judgment should always begin at home and that the
fundamental business of the moralist is to determine what
he himself ought to do rather than with what others ought
to have done. Again, Popper is opposed to essentialistic
theories of human nature which conceal an implicit and
uncritical normative claim within a pretentious "real" defini-
tion that relieves the individual of all responsibility in de-
ciding what sort of person he ought to become. But surely it
was the great historicists of the nineteenth century who pre-
pared the way for Popper's position by insisting that man,
the only being with a history, is therefore alone the being
without an essence. And surely it was the historical idealism

of Fichte and Hegel which first insisted that man's existence
as a moral being begins with the recognition that no law,
whether physical, logical, or even historical, can determine
what any spiritual being is to do or what it is to become. Un-
like Marx, Hegel did not profess to be able to predict, through
his dialectic, the next spiritual development of man. And he
did not do so precisely because there exists a domain of
"absolute" spirit, which is also a domain of absolute freedom
and responsibility, within which everything turns on what the
self at last recognizes itself to be. Here nothing is, or can be,
preordained, for here the spirit enters (or re-enters) history
of its own accord, and by itsjudgment determines what
shall be. The language, of course, is not Popper's, but it can
be translated, without too much difficulty, into a doctrine not
so very different from his own. And the fact that in his old age
Hegel's own absolute spirit turned out to be a functionary of
the Prussian state no more blurs this philosophical agreement

than does Popper's libertarian antipathy to the aged Hegel.
In essence, both preach a moral gospel of continent self-
determination which has nothing to do either with the laws
of nature or the laws of men.
From Popper's point of view the historicist is an equivoca-
tor who misled by a false logic into a nerveless and cynical
is
acquiescence in things as they are. But there is another

side of the, picture of historicism which Popper wholly ig-
nores and which, more sympathetically described, is not so
very different from his own best view. For example, it is
characteristic of the historicist to urge great restraint in judg-

ing "lesser breeds without the law," both on the ground that
objective historical facts about human actions are hard to
come by and on the ground that judgments which assign
responsibilities without regard to the principles avowed by
the lesser breeds themselves not only beg a point which may
be at issue ( what really is right? )and more impor-
but also,
tant, violate the principle that no man shall be judged mor-
ally save by a company of his peers. It is indeed worse than
pointless to blame men for the performance of actions of
which we disapprove on principle but which, on their princi-
ples, are either permissible or even proper; it is itself, from
our own point of view, immoral. Any significant moral dis-
agreement presupposes some sort of underlying spiritual
consensus. In its absence, we simply debase the currency
of moral discourse when we charge others with obligations
and responsibilities which their own principles do not per-
mit them to assume. The point here is not that, since every
one in fact views moral problems from a certain historical
perspective, we can never truly say that anything is either
right or wrong; it is rather that the freedom proper to any
moral agent entitles him to act on his own principles, and
that when we blame him for actions that run afoul of princi-
ples which he does not share, we ourselves may properly be
blamed for an insensitivity in the use of the language of
morals which itself amounts to a kind of moral fault. It ap-
pears to me that, as he makes it, Popper's own principle,
"Judge not!", which he does not always consistently honor,

has much the same point in view. The trouble is that it is
made to depend upon a supposedly logical theory that there
can be no such thing as moral objectivity. Since for him, as
for the positivists, there is no conceivable methodology for
their justification or verification, every moral judgment can
only be understood as an expression of some subjective senti-
ment or attitude, with respect to which questions of propriety
and validity cannot even arise. One wonders therefore
whether the "moral" freedom which Popper preaches is itself
anything more than a logical consequence of a meta-ethical
subjectivism which, in viewing all moral judgments and
principles as "man-made," converts the principle, "Judge
not!" into a kind of truncated logical truth. On this score,
one may again prefer the position of the historicists who at
any rate still regard self-righteousness as a moral rather
than a merely logical fault.
The most important philosophical use of the "historical
consciousness," of which Hegel made so much, is not only
that it serves to sharpen one's wits — on this score, any com-
—
parative study of cultures may do as well but also that, by
enlarging one's sympathies, it helps to make one less paro-
chial in one's judgments and more disposed to respect the
dignity and integrity of "lesser breeds" who, however be-
nighted their customs may seem to us, follow them as seri-
ously, as loyally, and as tragically as we do our own. Un-
derstanding the historical conditions which provide the con-
text within which all moral sentiments are acquired, one is
gradually made aware of the pointlessness of efforts at morale
reform which do not reckon with all of the complex circum-
stances which affect moral education and re-education. /
Aristotle wisely said that knowledge of the good is impossi-

ble without a sound moral training. In one way, the histori-
cists reaffirm and amplify that insight by showing how rare
are the conditions which make for such training and hence
for such knowledge. In so doing they also show us how
futile is that nostalgia for the golden ages of virtue, when the
knowledge of good and evil was securely possessed by sim-
pie, dutiful, god-fearing men. Living at the end of history,

the education of the historicist, like his moral problems, is
bound to be more complex. For while he may appreciate the

moral grandeur of the Prophets or the wise men of Greece, he
also knows that such simplicity and such unity of feeling are
not for him. Perhaps he may still regard the Ten Command-
ments as exemplary, but for him their proper meaning and
application are no longer self-evident. He sees why it is no
good simply to mumble over to oneself the First Command-
ment when one feels in one's heart that God is dead, and
why it is no good saying to oneself that one should honor
one's parents when one no longer has a spiritual home. In a
word, the lesson of both Hegel and Nietzsche is that if one
lives at the end of history, that very fact is a fundamental
condition of one's moral consciousness or existence, which
makes it necessary to face the bitter remoteness of all an-
cestral moralities. Hence again the "grey on grey," and hence
also the necessity of a philosophical-ethical reconstruction
which is made all the more difficult by the fact that the
parts out of which the reconstruction has to be made tend

maddeningly to go to pieces in one's hands.
One should face facts, moral as well as otherwise. But I at
least cannot follow Popper in his wholesale condemnation of

the historicists since I myself share many of the moral am-
biguities and perplexities which afflicted them. No more than
anyone else can I "make" a moral law for myself, as Popper
has evidently tried to do, and this for the reason that I find
all talk of such laws is as dead as the God in whose name
they have historically been proclaimed. For me, also, it is not
a question of deciding what to do, but of first discovering
morally what I am and what my principles really are. Many
of Popper's own libertarian precepts strike a responsive

chord in my own conscience. But the sound is muffled, for I
hear also the cries of those for whom the only fundamental
moral reality is estrangement. How then shall I condemn all
of the ideals of the closed society in the names of liberty and
equality, so long as the ideals of the open society neglect the
demands of community and fraternity? And how shall I con-
demn all of the tribal "magic" of the institutions of com-

munity when, in its absence, what an aggregation of
I find is
moral atoms, each as lonely as it is autonomous, and equal
in nothing but its sense of isolation? Popper still lives

spiritually in the aftermath of the American and French
revolutions. He is a true son of Paine and Jefferson and the
French philosophs, and, up to a point, I honor him for it. But
their world, alas, is not my world, any more than it is the
world of Marx or Nietzsche or Freud. And it is for this reason
that however much I may deplore or even loathe the moral
standards of the totalitarians, I cannot so unqualifiedly give
my allegiance to the "open" society as Popper has described
it. For what he neglects is the history of the nineteenth
century which, for profoundly important reasons, could no

longer simply reiterate the ideals of the Enlightenment.
There is one final aspect of Popper's moral philosophy
which requires some comment. As I have already remarked,
although Popper is a philosophical analyst who takes very
seriously the study of the logic of scientific inquiry, he de-
plores the "useless" preoccupation of other analysts with the
problems of defining ethical terms. The fact is, however,
that his own conviction that all discussions about definitions
of such terms as "good" and "right" are useless itself pre-
supposes an analysis of their meaning or use which precludes
the possibility of defining them in terms that are used to
describe the objective, empirical characteristics of things.
Popper thus agrees with the intuitionists that any attempt to
construct an empirical science of ethics is doomed from the
outset by the impossibility of giving its basic concepts a
genuinely empirical meaning. Let us agree to this. Let us
also agree with him ( and Moore ) for the sake of argument,
,
that all attempts to define goodness have proved futile. It

still does not follow that they are useless, since it was only
after repeated attempts to define goodness that anyone was
disposed to explain why it is indefinable. What I object to is
Popper's assumption that since the truth about ethics has

been proclaimed, it is idle for anyone else to raise questions
concerning the meanings of ethical terms, and that hence-
forth the logical analysis of moral discourse is an escape from

the realities of the moral life. Indeed, he is not even consis-
tent, for a great part of his own critique of historicism is
predicated on the thesis that incorrect analyses of moral

thought have momentous practical consequences and that
a sound analysis, such as his own, is a corrective to many of
the spiritual confusions to which the proponents of the
closed society are heir.
What I fundamentally object to in all this is what I also
complain of in the case of the positivists, from whom Popper
differs technically but not ideologically, namely, the tradi-
tional philosophical rationalism which assumes that what is
not a problem of knowledge ( as rationalism elects to define
knowledge) is not a problem at all, and that since morality
is not (in the preferred sense in question) verifiable, every
moral issue is resolvable only by a decision. The odd con-
sequence is that, although the processes of reasoning by
which the two schools arrive at their conclusions seem so
widely different, the rationalists are led, slowly but inevit-
ably, to the existentialist conclusion that at bottom moral-
ity, like religion, is intellectually absurd. Thus, to the precept,
"Judge not!" we may add another: "Don't stand there; do
something!" But what if one decides, absurdly, that the thing
to do is to think a little more about the nature of morality it-
self?
POSTSCRIPT ON SOCIALISM
In conclusion let me briefly summarize what seems to me

good and bad in Popper's general position. In the first place,
in some ultimate sense, I am inclined to side with his generic
point of view, from which stems much of his antipathy to
Plato, Hegel, and, in part, Marx. So far as it goes, this point
of view may be characterized positively as (a) scientific
and naturalistic, but not narrowly positivistic, in historical
methodology; (b) humanitarian and liberal in morals and
in politics. Negatively it is (a) anti-intuitionistic in logic and
in philosophy; (b) anti-authoritarian in ethics, and (c) anti-
totalitarian in politicsand ideology. All this I associate witfo

sanity and good will. no reason simply because of
But, I see
this to reject limited, gradualistic, democratic socialism. No-
where does Popper show that free society is incompatible
with social organization of modes of production or even dis-
tribution, provided that constitutional government, due
process of law, and the preservation of essential civil liber-
ties, including freedom of press and speech, are rigorously
adhered to. In countries like Great Britain, the basic con-

stitutional and political structures may survive profound
changes, involving greatly increased state control of social
and economic institutions, without jeopardy to the basic
democratic political controls which insure society against
and unchecked abuses of governmental authority.
willful
Finally,Popper seems to me to have evaded the problem
which in two-thirds of the world is most pressing today:
What shall be done by the men of good will who confront
entrenched tribalistic closed societies in their own countries?
Piecemeal social engineering is all very well for democratic
countries which are already partially "open." But such a
choice is not open to the people of many countries. It would
be refreshing, at any rate, to find in Popper any echo of
Thomas Jefferson's affirmation of the eternal right of a people
to revolution against tyranny and oppression.
It is not without significance that the emphasis in Popper's
book is upon a maintenance of present institutions in the
"democratic countries" rather than upon the radical changes
which many of us believe are essential in non-democratic
countries. Again and again he praises the improvements
(which Marx fully admitted) that have occurred in the
condition of the working classes under capitalism. He points
out the high standard of living prevailing in the United
States, due to the free-enterprise system. He does not men-
tion the misery, wretchedness, and grinding poverty that
also exist; nor does he point out that many of the improve-
ments are due and the increas-
to the organization of labor
ingly effective use of its economic power. He praises those
who "emphasize the tremendous benefit to be derived from
the mechanism of free markets, and who conclude from

this that a truly free labor market would be of the greatest
benefit to all concerned." 2
He forgets that a "truly" ( essen-
tially?) free labormarket has in the past often resulted in
exploitation and "economic terrorism." The point is that it is
impossible to remedy the ills of laissez-faire without
limiting someone's freedom. Either you increase the power
of government to interfere in economic arrangements and so
impede the free private use of economic power (by capital
and labor), in which case you approach socialism in fact,
whatever you call it in name, or you allow the "free"
use of economic power, in which case you get large-scale
monopoly capitalism and a closed labor market.
In the end Popper is merely equivocal. A more candid
facing up to the problem of power and the inescapable neces-
sity of its increased and responsible use by the state, if the
social and economic conflicts of modern civilization are to
be resolved, would have led him to assume a far different
position from that which he adopts in this book. He would
then have seen, perhaps, that some of the apparent enemies
of the "open" society whom he so bitterly excoriates are its
real friends in a common pursuit of "the wisest" and "the
best."
2 Popper, op. cit., Vol. II, p. 116.
[XIV]
Utilitarianism and Liberty:
John Stuart Mill's
Defense of Freedom
MILL AND THE ESSAY "ON SOCIAL FREEDOM"

In 1941 there appeared a volume bearing the title On Social
Freedom, the authorship of which was unequivocally attrib-
uted by its editor, Miss Dorothy Fosdick, to John Stuart
Mill.The work had originally appeared in 1907 in the Ox-
ford and Cambridge Review, the original manuscript hav-
ing been found after Mill's death among his other papers.
More recently, Mr. J. C. Rees has argued convincingly that
Mill cannot be the author of the essay, and I gladly accept
his verdict. The whole style of the work, with its rather
heavy-handed jocosity, its poor organization, and its close
affinities with the schools of intuitionism and idealism, always
made it a rather puzzling document for those who, like my-

self, had no reason to question Miss Fosdick's account of the
matter. Had I been disposed to doubt Mill's authorship on

the basis of internal evidence, the editor's acknowledgment
to Professor R. M. Maclver for his "encouragement and ad-
vice," together with the fact that the volume was pub-
lished by the Columbia University Press, would have suf-
ficed to reassure me. But no such doubts occurred to me, nor
did I ever hear any expressed, until it was pointed out to me,
after the general subject of this paper had already been pro-
posed, that Mr. Rees had established beyond peradventure
that my easy assurances were without foundation.
Mill is an interesting person, and any work ascribed to him
must be taken seriously, even by one who, like myself, has
been primarily interested in his ideas themselves, rather
than in his advocacy of them. The issues raised in the essay
"On Social Freedom" were, and are, serious ones. This essay,
which I accepted as Mill's, forced me to re-examine his treat-
ment of the issues in the essay "On Liberty," and to dis-
cover in consequence how little I could agree with his de-
fense of social liberty. The fact that "On Social Freedom"
does not do full justice to some of its own implicit criticisms
of Mill made it all the more desirable that others should
seek to do so. Let me present the situation in the bluntest
possible terms : nearly everyone, at least in this country, pays
lip service to the depth of Mill's feeling in the essay "On
Liberty"; yet nearly everyone acknowledges his argument to
be defective. What, then, is there left in the essay to com-
mend? It is not unfair, I think, to say that the predicament
of most contemporary moral and social philosophers in re-
gard to the essay "On Liberty" is similar to the position of
/"-many contemporary theologians in regard to Christianity;
I that is to say, they try, albeit unsuccessfully, to swallow
\ Mill's argument for the sake of its sentiment, yet, because of
<their unavoidable doubts about the argument, they cannot
/help being secretly doubtful of the justifiability of the
V sentiment itself. And it is precisely for this reason that con-
temporary liberalism, caught between a commitment to the
absolute value of liberty and a utilitarian belief that all liber-
ties can be justified only by an appeal to something called

"the general happiness" or "the general welfare," pass
uneasily back and forth between a last-ditch defense of
liberties whose general utility remains unproved and an over-
ready acceptance of their curtailment in the supposed in-
terest of the common good.
My own view is that we must fish or cut bait: if Mill's
Utilitarianism and Liberty 295
conception of liberty as well as his defense of it will not

stand, as the author of the essay "On Social Freedom" im-
plies,then we must ask whether his libertarian sentiments
should not be rejected, as the Marxists, neo-Thomists, and
idealists maintain. If,on the other hand, we regard those
sentiments as among
the most precious of the attitudes
which compose our so-called way of life, then we must
question whether the principle of utility, as their covering
principle of moral and social justification, must not be flatly
given up. In that case, we will have moved away not only
from one of the positions of the essay "On Social Freedom,"
which may be no great loss, but from the position of the
essay "On Liberty" as well. My taste is to take the latter al-
ternative, even if it involves a break with the Benthamite
tradition of philosophical radicalism as sharp as the break
of the philosophical radicals with the doctrines of natural
law and the social contract. Mill himself was unable to make
such a break and it is precisely because of this that an im-
partial rereading of "On Liberty" is such a disillusioning ex-
perience.
The merit of the essay "On Freedom" is that it
Social
forces us at once to reconsider Mill's whole individualistic
conception of liberty, to which it is opposed, and to ques-
tion the general utilitarian defense of social freedom, which
it accepts. Let me be more explicit. In the first place, by
taking a more consistently utilitarian stand in regard to the

appraisal of liberty, the author of that essay is able to show
without too much difficulty how far, on such grounds, the
defense of individualism is compromised from the outset.
From such a point of view, the liberty of individual persons
cannot be regarded, as Mill tends at times to do, as an end
in itself. For the utilitarian, liberty is, at best, a means, not
an end, and the defense of any particular liberty must be
solely on the ground of its social utility. Mill himself had
argued for the principle that "the sole end for which man-
kind are warranted, individually or collectively, in interfering
with the liberty of action of any of their number, is self-
protection." 1
This may be an admirable precept, but the
case for it, on purely utilitarian grounds, is extremely dubi-
ous. On
such grounds, indeed, there can be no theoretical
limit whatever to the restrictions of individual freedom of
thought and action. And if, in practice, as Mill himself ad-
mits at times, the majority of mankind is neither very wise
nor very prudent, then it may be necessary for their wiser
and more public-spirited utilitarian leaders rigorously to reg-
ulate their activities both for their own sakes as well as for
the common good. But in the second place, as the essay "On
Social Freedom" implies, Mill's theoretical consistency is
hardly less defective than his sense of fact. No sharp dis-

tinctions can be made, as Mill supposes, between spheres of
private activity within which the individual person is alone
concerned and a sphere of public interest which is the
exclusive interest of something called "society." In one way
or another, virtually every human activity is other-regarding,
just as, in one way or another, it is also other-affecting. If,
then, the defense of privacy rests on nothing more certain
than the thesis that there are affairs of men which are essen-
tially non-social and in which, therefore, society as a whole
has no legitimate interest, then not even the affairs of the
toilet can be properly regarded as matters of exclusively
private concern. Virtually every human activity, not only in
the cultural but also in the material and economic sphere,
depends upon the active cooperation of others, in many cases
not merely as individuals but also as groups. To that ex-
tent, every human activity affects the interests of others,
and, as such, is a matter of public concern. Hence, if the de-
fense of non-interference in a sphere of activity is simply
that the activity is self-interested, then nowhere can, and
perhaps nowhere should, that defense succeed. If, on the
other hand, its defense depends upon the supposition that
a sphere of activity does not involve the well-being of others,
and it therefore is of no proper concern to them, the de-
fense equally fails. In either case, Mill's case for liberty
X
J. S. Mill, On Liberty (London: J.
M. Dent & Sons, 1944), pp.
72-3-
is doomed. To be sure, the author of the essay "On Social

Freedom" does not quite manage to make this point; yet he
says enough to allow others to draw the inference for them-
selves. And if somehow we remain unconvinced by such an
argument, then it is not so much because we really believe
the facts to be otherwise, but because we realize at last that
the premises of the utilitarian ethic must be disallowed. Un-
fortunately, however, this is precisely what Mill, who ex-
plicitly rejects the Lockean view of liberty as an "abstract
right, as a thing independent of utility," 2 is unable to do. As
he still forces himself to say, even in the essay "On Liberty"
itself, "I regard utility as the ultimate appeal on all ethical
3
questions." The commitment is, to be sure,
flatness of this
seriously qualified by the proviso that "utility" is to be con-
strued "in the largest sense, grounded on the permanent in-
terests of a man as a progressive being." 4 But this means
merely that Mill himself can accept the principle of utility
only if he is allowed to define what should be meant by "the
public welfare," or, better, the individual interests in terms of
which the public welfare is to be constituted. How far such
normative qualifications are removed from the concept of
"the greatest happiness of the greatest number," as under-
stood by Jeremy Bentham, the founder of utilitarianism,
needs no emphasis. Neither Mill nor the author of "On Social
Freedom" is remotely a Benthamite. Yet the latter, whatever
his qualifications of the original utilitarian standard may
come to, really does mean to judge liberties solely by their
social consequences; nor is he in the least dismayed by the
strenuously limiting judgments which he is obliged to make.
Mill, on the other hand, both loosens the utilitarian standard

itself and overhauls the facts of long-run private and pub-
lic interest so as to guarantee in advance that, at any cost, the
standard will be left intact. He is immensely imaginative,

not to say fanciful, in construing individual liberties as pub-
lic benefits. He is incredibly dull when it is a matter of search-
2 Ibid.,
p. 74.
s Ibid.
4 Ibid.
ing out, impartially, reasons to limit liberties whose exercise,

on his own grounds, may result in public misfortune.
We must, then, be grateful to the author of the essay "On
Social Freedom" for forcing us to reconsider the whole basis
of the modern liberal defense of liberty. And if he leaves us
in the lurch at all crucial points, he at any rate says enough
to show that any true disciple of Mill has his work cut out
for him. In what follows, I shall first discuss critically in
some detail certain important similarities and differences be-
tween the essays "On Liberty" and "On Social Freedom." In
the concluding part of this paper, I shall offer in summary
form the sketch of a very different view of liberty than is to
be found in either of these essays. Briefly, I consider the case
for liberty, either on strict utilitarian or on idealist-utilitarian
grounds, very shaky indeed. At one point or another, both of
these points of view betray the cause of liberty; in the case
of the idealists, the betrayal was more or less explicit; in the
case of the utilitarians, it remained for the most part implicit,
and, as in the case of Mill, recognition of the fact was
forestalled only by muddle-headedness and inconsistency.
Unfortunately, I am not much attracted to those aspects of
Mill's own defense of liberty which, as many critics have ob-
served, verge away from utilitarianism. In my opinion, they
add confusion without appreciably increasing the strength of
his position. Specifically, Iam not impressed by the thesis
that, for most men, liberty forms the great part of happiness.
Erich Fromm, among others, has shown how dubious is this
thesis as a general proposition. Still less am I persuaded that
much of a case can be made for liberty on the ground that it
is the condition of individuality. Individuality is a variable
commodity of whose value Mill made a fetish. Nonetheless,
liberty itself is a great good, so great in fact, that it ought
not to be jeopardized by such precarious defenses as these.
similarities between on liberty and

"on social freedom"
It will be convenient first to discuss critically certain points of
similarity between the two essays, and then to consider
some of their differences.
As to their similarities, it is important to remark at the
outset that although both essays are ostensibly limited to
the topic of social liberty or freedom, neither entirely suc-
ceeds in confining itself within these limits. The essay "On
Liberty" contains an elaborate discussion of the ethical ideal
of individuality, much of which is only indirectly related
to the problem of social freedom. The essay "On Social
Freedom" contains remarks upon the general question of lib-
erty and necessity and some closing comments upon the
doctrine of motives, including the concept of "higher" and
"lower" motives, whose bearing upon the problem of social
freedom is not made apparent. The proper implication to
be drawn, however, is that the question of social freedom is
not an isolatable topic; proper discussion of it as we shall see
raises virtually every question that can be raised concern-
ing the doctrine of liberty or freedom. Nevertheless, the
manner in which it is discussed precludes adequate treatment
of certain important topics, and for this reason it is necessary
to remark upon the formal limits which both authors set for
themselves. By social liberty or freedom both of them under-
stand pretty much the same thing: that is, the liberty of
individual men in relation to the constraining power of
something called "society." And what interests them are the
nature and limits of the constraining power legitimately ex-
ercised by society over the individual, and, conversely, the
proper limitations upon individual freedom by society. Mill
is more interested in the limitations of social control; the
author of"On Social Freedom" is more interested in the limi-
tationsupon individual freedom; but the difference is mainly
a matter of emphasis. The important thing is that neither
writer is much concerned with the freedom of particular
societies in relation to other societies, or the freedom of

societies in relation to the restraining power of individuals.
Neither carefully considers liberty as a freedom from en-
vironmental necessitations and determinations of behavior
of which political compulsion and legal constraints are
merely particular forms. Both formally acknowledge that
other such constraints exist, and the author of the essay "On
Social Freedom" tells us in effect that power of law may be
useful as a way of offsetting other forms of social constraint.
But neither discusses in any detail the ways in which non-
politicalforms of organization may serve to protect the
individual against the encroachments of government and to
assist the individual inremoving burdensome laws or in
changing the law with a view to more effective forms of
social action which, in practice, may serve to guarantee
social rights. More significantly, perhaps, neither discusses the
many ways in which, as the idealists were fond of pointing
out, nonpolitical as well as political institutions constrict
the area within which individual freedom is exercised —by
formal education, by moral training, and, in general, by the
various informal social determinants of character. Thus nei-
ther considers, for example, whether parents should be free
to determine the character and attitudes of their children
which, even after the latter come to the age of maturity, still
effectively limit the range of their deliberations and choices
in many important spheres. In the essay "On Liberty," Mill
expressly says that his doctrine only concerns mature persons
who "have attained the capacity of being guided to their
own improvement by conviction or persuasion." 5 In fact, so
far in this direction does he go that he argues that "Despot-
ism is a legitimate mode government in dealing with bar-
of
6
barians. . .
." But what if men like to be "barbarians" and
want to train their offspring to prefer barbarian habits and
customs?
These problems become even more exigent when we ask
directly about the limits of moral freedom. Given certain
5 Ibid.,
p. 73.
e Ibid.
notions of religion, it may perhaps be argued that the reli-

gious life concerns nothing more than a man's relations to his
own God, a matter which, to that extent, has no direct bear-
ing upon the common good. But every form of morality,
utilitarian or otherwise, isconcerned with the rights and re-
sponsibilities of men toward other men. On the other hand,
there is no morality unl ess men are free not only to do as
they please but also as they think they ought. Where mo-
rality takes the form of law, whether it be in the name either
of the greatest happiness or of a so-called law of nature,
moral autonomy so far disappears. In that case, even if our
remaining liberties amount
virtually to license, freedom of
conscience — from freedom in the pursuit of scien-
as distinct
tific truth and freedom of worship —
no longer exists. Now up
to a point, of course, Mill himself defends moral freedom,
but only as part of that freedom of thought whose aim is
what is involved
the pursuit of objective truth or as part of
in self -development.Nowhere, so far as I can see, is moral
freedom discussed on its own grounds as the condition of a
form of life whose value to us derives from no other source
whatever.
In the second place, neither essay gives much attention to
the concept of society. However, it is evident upon analysis
that what is meant in both essays by "society" is, on the
whole, a politically organized society of the sort exemplified
by a modern nation-state. For neither author is a form of non-
political association such as a trade union, an international
social class, or humanity as a whole considered a form of so-
ciety — or, perhaps more important, as a social institution.
This point is of importance in a number of ways. For ex-
ample, Mill admits, more or less in passing, that individuals
may be constrained not only when their actions would be
harmful to others, but also in some cases when the perform-
ance of the act is a positive benefit, or the non-performance
an injury, to them. Thus, a man may be rightfully compelled
to give evidence in a court of law, bear his fair share in the
common defense, or "in any other joint work necessary to
the interest of the society of which he enjoys the protec-
7
tion." But of course such forms of constraint mean one thing
if one talking about a nation-state, another if one is talking
is
about some other form of social organization. A Marxist,

for example, might well argue that forms of compulsion
which Mill countenances without batting an eye are in the
highest degree arbitrary and illegitimate interference with
the social freedom of the international working classes. Or
a Christian might argue that compulsory armed service for
the defense of an un-Christian state is a form of tyranny
which ought to be resisted. The fact that Mill and the
author of "On Social Freedom" tend to identify society as
such with the nation-state thus unconsciously puts a pre-
mium upon forms of coercion that are necessary to the pres-
ervation of welfare of nation-states. Liberties and restraints,
arid hence rights and duties, that cut cleanly across political
and legal boundaries are thus scarcely acknowledged to
exist. To be sure, both essays acknowledge such forms of
restraint as may arise from the pressure of custom and public
opinion, and the author of "On Social Freedom," at least, is
aware of some of the ways in which economic sanctions may
affect the use of government and law as ways of protecting
the individual against the coercive and restrictive power of
non-political organizations within a politically organized
he nor Mill sees how the countervailing
society. Yet neither
power of non-legal associations and nonpolitical institutions,
both within and without the boundaries of nation-states, may
serve as shields for the defense of common social liberties
against the abuses of state power and the many subtle forms
of oppression that are so often sanctified in the name of law.
In the modern world, it is frequently not the defense of
individual liberties against the state which needs arguing, but
the principle of institutional autonomy. Most liberals accept
Holmes's dictum that individual civil liberties should be
qualified or sacrificed only in the case of a "clear and present
danger"; they are also prone to extend it to some, although
not forms of cultural organizations such as education and

all,
the church. It should be borne in mind, however, that

7 Ibid.,
p. 74.
such a view still accepts clear and present dangers to the na-
tion-state as primary. Such a position may be defensible
when one is Holmes was, simply as a judge and
arguing, as
jurist. In that context, no doubt, the position may pass as "lib-
eral." It becomes far more dubious when treated, as most

liberals do, as a moral principle. For, so construed, it in effect
places the state in a position of peremptory privilege, and,
in times of political crisis, requires not only the sacrifice of
the civil liberties of individuals but also the social freedom
or autonomy of other, perhaps higher institutions. Mill's
position unconsciously anticipates that of the followers of
Holmes. And it does so precisely because the only ultimate
form of liberty which he acknowledges is the liberty of indi-
viduals and because the only form of society which he seri-
ously contemplates is that of the nation-state. Had he con-
ceived of the concept of liberty as applying not only to
individuals, but, without reduction, to groups, and had he
conceived society in other than purely legal and political
terms, he might well have seen many of the problems of so-
cial freedom in an entirely different light. And it is because of
this, I fear, that the defense of the utilitarian ideal of the
general happiness as well as that of social liberty or free-
dom has passed, for so many decades, from his followers to
men whose conception of the good life, in the end, is very
remote from Mill's own.
DIFFERENCES BETWEEN THE ESSAYS

Let me now examine certain major differences between the
two essays which will enable me to underline the moral
which I subsequently wish to draw from this discussion.
In the essay "On Liberty," Mill adopts a conception of
self-interest and of the individual's power to satisfy it which
differs only in degree from that of Locke. And he tends to
think of the collective interest or "welfare" of society as an
aggregation of the self-interests of its members. Finally, al-
though he rejects the Lockean theory of the social contract,

he, like Locke, nevertheless tends to think of civil society
and its institutions as existing solely in order to protect the

self-interested activities of individuals from interferences due
to the self-interested activities of other individuals or groups.
The plain implication is that to the extent that an individual is
capable of satisfying his self-interest by his own efforts he

should not be restrained by "society" except in so far as his
actions interfere with the actions of others in their own be-
half. From this standpoint, the virtue of collective action is
mainly the negative one of protecting the individual from
overt harm due to other individuals and to prevent him, in
turn, from harming others in the pursuit of their own
As in the case of Locke, Mill also
self-interested activities.
explicitly rejects the view that an individual can "rightfully
be compelled to do or forebear because it will be better
for him to do so, because it will make him happier, because,
in the opinions of others, to do so would be wise, or even
right." 8
The essay "On Social Freedom" represents, on the whole,
a quite different point of view. Partly because of its brevity,
it does not press its argument very far; but there can be no
doubt that it regards the basis of Mill's distinction between

the domains of private and public interest as untenable. In
the earlier essay, private interest is largely taken to coin-
cide with "self-interest," and "public interest" simply with
the aggregation of self-interests of the members of a society.
As the author of "On Social Freedom" argues, however, self-
interest is frequently not limited to what we commonly re-
gard as the domain of private activity. What a man does in
private life is largely determined by his view of what other
people will think of him. The love of money, the love of
property, is as much affected by the social passion for emula-
tion as by the interest in personal security. In short, what
men aim at, in their own interest, are not merely the "neces-
sities of life," but social position, power over their fellows,
and, in general, the goods represented by the term "status."

But the matter goes still deeper than this. Now in
Utilitarianism Mill himself had made a great point of the
8 Ibid.,
p. 73.
role of social feeling in the moral life. But he did so mainly

lSTbrder to provide a basis for the principle of utility other
than the prudential self-interest which, for the elder utilitari-
ans, had provided the fundamental motive for altruistic ac-

tion. In short, Mill introduces social feeling or sympathy
partly in order to protect utilitarianism against the charge,
which Butler had earlier leveled against Hobbes, that it re-
duces the whole moral life to a calculus of pleasures or in-
terests whose principles or laws are nothing more than
summary prudential rules which at once lose their authority
when the conditions of personal well-being are altered. No-
where, however, does Mill lay much stress upon the more
positive roles of the non-moral fraternal and communal im-
pulses in human nature. On the contrary, so far as he attends
to them at all, he tends to view them primarily as tendencies
toward conformity which inhibit the individual's capacity
for free self-development. And it is significant, I think, that
his whole approach presupposes that he views what may be
called the socialization process as a factor which merely
serves to constrict the individual's freedom to develop his
own personality. But is this so? Is it not true, rather, that for
many men spiritual self-development means, in effect, sub-
mersion of the private self in some great collective enterprise
or work? And for them may it not be the freedom of collec-
tive interests and actions against the hampering activities of
egoistic individualists that constitutes the most serious prob-
lem of social freedom?
I mention this possibility not because I delight to contem-
plate it, but because it is necessary to underline the point

that Mill, like most men, is really interested only in a certain
kind of self-development, rather than in self-development
as such, and because he so little appreciates the view, repre-
sented by the idealists, that if self-development is the pri-
mary element of man's spiritual well-being, then social life
cannot, in general, be regarded simply as a threat to in-
dividual freedom. On this score, at least, the author of "On
Social Freedom" comes somewhat closer to seeing that,
from a moral point of view, the essential problem may be
not whether society as such has fairly got the better of in-
dividuality, but rather what sort of individuality is worth

cultivating, and whether a particular social institution or
system puts that sort of individuality in jeopardy.
There is one important respect in which, surprisingly
enough, the argument of the essay "On Social Freedom" is
more characteristically utilitarian than that of the essay "On
Liberty" itself. In the latter, Mill sometimes speaks as though
the very idea or feeling of being unconstrained is itself an
intrinsic part of individual happiness and hence of the
general welfare. If I correctly interpret him, he maintains
that not merely is doing what you please to do a good, but
the idea that you can do whatever you might please to do
and the sense of spontaneity are also positive goods in their
own right. In this sense they are essential ingredients or
parts of the general welfare which the principle of utility
envisages as end and which it is the primary business of
its
society and government to protect. The author of the essay

"On Social Freedom" appears to reject such a view. He is
concerned with the values neither of spontaneity nor of the
idea of liberty. For him, I gather, liberty is in every sense
merely a means to an end. Liberty is desirable, when it is so,
merely because without it some particular good would be
unrealizable. From his point of view, the general principle
that every man
ought to be at liberty is, as it stands, absurd.
Again, there is no liberty that is not a liberty to do or not to
do something in particular. It is worth defending or imple-
menting only if the end of which it is a condition is ante-
cedently desirable. Therefore, liberty, forms no essential part
of the general welfare conceived as an end. It is desirable,
when it is so, only as a condition of the end, and then only in
so far as that end is itself desirable.
Here, as seems to me, Mill is partly right, but for the
it
wrong reasons. I have no doubt that the idea of being able

to do as one pleases is, for many men, a great good. But
that idea is not itself a form of freedom. Likewise, the sheer
feeling of being unconstrained or the sense that one is act-
ing spontaneously, as Hume had earlier pointed out, is, as
such, not a form of liberty. On the contrary, it may easily be

present in forms of behavior in which the individual is acting,
without thought or choice, merely at the behest of another
person. If it is a good, and I do not doubt that it is, it is only
one good among many. Nor is there any clear evidence that
it will occur more frequently in an individualistic society
than in a collectivistic society.
For my part, however, it is not the idea of being free or the
sense of being free, but freedom itself that principally mat-
ters when we talk of rights and responsibilities. A man
ought to be at even though in being so he may have
liberty,
a smaller share of the feeling of unconstraint than have men
who live in a smoothly functioning, planned society.
I also think that Mill supposes there to be a far more inti-
mate connection between political and social liberty and in-
dividuality than in fact exists. In our country there is con-
siderable political liberty, but far less individuality than in
many which there is a greater amount of politi-
countries in
But I do not believe that this fact, if such it be,
cal constraint.
would provide a good reason for curtailing or minimizing the
value of the liberties we presently possess. If individuality
is a good,must be cultivated by other means. Mill, the son
it
of a highly tyrannical father and the victim of one of the

most remorseless educations to which a precocious youth
has ever been submitted, could only conceive of freedom as
a sphere of private activity within which the individual can
simply "be himself." On this score, the author of the essay
"On Social Freedom" occupies a firmer position. He quite
properly attacks the whole strategy of those liberals who
seek at the same time to confine men's legitimate aspirations
toward freedom within a Chinese wall of individualistic pri-
vate activity which corresponds neither to the social and
political liberties to which most men aspire nor to the long-
range happiness of society itself. And he sees also that if
liberty is a good, it is a good for conformists and nonconform-
ists alike. What he does not see so clearly is that, regard-
less of the sphere within which it is exercised, and regardless

of its effects upon the general welfare, the liberty of men is
an intrinsicmoral good. No human good, whether it be

liberty or life itself,can be defended without qualification.
But a beginning in the right direction can be made only by
the firm adoption of the principle that every person, simply
as such, has a right to be at liberty.
This brings me to a final but striking theoretical difference
between the two essays. In the essay "On Liberty," Mill
has nothing at all to say abut the so-called metaphysical
problem of freedom versus determinism. I presume that the
reason he did not discuss it in that work is that, like Hume,
he felt that it forms no part of the problem of moral or social
freedom. We know from his other works, however, that as
a philosopher of science and of history Mill was a strict de-
terminist. On the other hand, the author of the essay "On
Social Freedom," in marked contrast to Mill, opens his an-
alysis with a perfunctory but none the less definite rejection
of determinism. Unlike Mill, he regards indeterminism as
an essential condition of a morally free will. He also appears
to think that the doctrine of determinism is incompatible
with any "practical assertion of the existence of individual
freedom."
Now when I first read this discussion, I was impressed
only by the slackness of the analysis, which seemed to me to
confuse a metaphysical or logical problem, indeed such if
a problem exists, with a purely moral and social problem.

From the time of Hume on, liberal moral and political phi-
losophers have generally insisted that the question of deter-
minism and the question of moral and social freedom have
nothing whatever to do with one another. I am now con-
vinced that the matter cannot be so simply disposed of, and
that, however inadequately he argues the case, the author
of the essay "On Social Freedom" was right in maintaining
that there is a genuine practical connection between the
metaphysical doctrine of determinism and the problem of
social freedom. Unfortunately, I must here state my case in
his defense in terms that are too brief and too dogmatic to
make immediately convincing to every reader. Let me
it
emphasize once more that in my judgment most metaphysi-

cal questions, including most so-called ontological ones, have

a practical and ideological as well as a purely logical aspect.
Now the doctrine of determinism forms no part of the posi-
tive content of any science or of any scientific description
of what there is. Positive science seeks only to formulate veri-
fiable hypotheses and theories about the behavior of classes
of phenomena under certain conditions. The cash value of
the metaphysical doctrine of determinism, I believe, is sim-
ply that of a procedural resolution to search unceasingly for
more and more adequate explanations of phenomena and
for more and more reliable hypotheses for prediction. Sup-
pose, however, that a given society is committed morally and
legally to a conception of spheres of free activity within
which not only no scientific experimentation but also no
scientific observation is permitted. Suppose, also, that the
society systematically and effectively succeeds in preventing
men from putting into effect the policies and procedures
that, in science, are necessary for the acquisition of certain
kinds of information concerning phenomena, and in partic-
ular concerning human phenomena. My contention is that in
such a society commitment to a deterministic philosophy
would be not so much false as pointless, and that for all
practical purposes acceptance of a certain element of irre-
ducible indeterminacy in the sphere of human activity would
be the only sensible attitude to take. Let me make the point
even more strongly: Suppose that the members of the sci-
entific community are themselves opposed on moral grounds
to certain forms of experimentation with and observation
of human behavior. Are they not on principle thereby com-
mitted to proceed as though there were certain unavoidable
restrictions upon the very freedom of scientific inquiry and
hence of the pursuit of knowledge and truth? And because
of this are they not correspondingly committed in practice to
assume that a certain indeterminacy in human conduct ought
on principle to be acknowledged? Indeed, does not this ob-
ligation, together with the corresponding right, imply in
effect that a certain commitment to indeterminism may be
part of any effective doctrine of social freedom which,
morally and legally, vouchsafes to the individual a certain

area of private activity within which no inquiring mind shall
be permitted to trespass? On the other hand, would not an
effective and consistent commitment to the doctrine of de-
terminism commit anyone who accepted it to reject any ham-
pering restrictions upon the free pursuit of knowledge and,
by a kind of irony, to the belief that the only sort of liberty
that ought to be defended is the freedom of inquiry which
Mill himself extols with such vigor? These questions are
complex. Nor do have the answers to them all. I raise them
I
only in order to show how problems of morals and meta-

physics may intersect and how shallow is the prevailing lib-
ertarian acceptance of metaphysical or scientific determin-
ism.
THE MORAL FOUNDATIONS OF LIBERTY

From what has now been said, it will be clear that I believe
that the utilitarian defense of liberty, like its defense of jus-
tice, is doomed. Morally, as Mill himself is constantly forced
in practice to admit, the problem seems always to be not
whether to grant or to extend a liberty as a right but whether
to disallow or to limit it. And this suggests to us that the utili-
tarians have fundamentally misconceived the ethical role of

the principle of utility itself. They have endeavored to treat
it as a supreme principle of justification for other sup-
posedly lower-order practices and principles. From such a

point of view, neither liberty nor justice or even individual
acts of kindness have any intrinsic merit. Nor can they
independently form the basis of any moral right or obligation.
I maintain, on the contrary, that the principle of utility, in
whatever form one may choose to state it, is at best to be

viewed as a principle for the making of exceptions to other
principles that are themselves independently binding. Lib-
erty is not a moral good only because it may conduce to the
general happiness, nor even because, by an extension of
^* the original meaning of that idea, it may form a part of the
V happiness of most men. Such a defense is wholly problematic
and leads those who employ it into unreal distinctions and

imponderable estimations. The moral foundation of liberty,
I contend, is nothing other than the right to be at liberty it-
self. In short, the f ountainhead of freedom ( if the phrase may
be allowed) is not utility but simply and solely the principle

that every person has a right to be at liberty. This principle,
I contend, requires justification by no other principle what-
ever. Nor does it require the support or sanction of any higher

authority, institutional or otherwise. It is no more to be
viewed as God-given, as a law of nature, or as a dictate of
pure practical reason than the principle of utility itself. What
authenticates it is merely our own conscientious avowal of
it. In the language of Kant, the principle of liberty is cate-
gorically imperative.
However, one or two possible sources of misconceptions
must be removed. In saying that every person ought to be
at liberty, I do not, like Mill, mean to restrict the principle
to every adult person, every reasonable person, or every
knowledgeable person; nor do I suppose that every person
must prove his right to be at liberty by proving his moral
competence, his humanity, his submission to the law of
the land, or his readiness to fight in the defense of freedom.
I make no such limitations of its application, and I deny that
any such limitations can conveniently be made. I mean just
what I say: "every person has a right to be at liberty." I use
the word "person" advisedly and deliberately. In the moral
sense, "every person" does not mean "every man," "every hu-
man being," "every sentient being," or simply "everybody."
Nor does it mean "every individual." Morally, as well as
legally,groups as well as individuals may be regarded or
treated as "persons." In the moral sense, the term "person"
may properly be applied to any individual (or group of in-
dividuals) toward whom a moral responsibility is due and
who is therefore in a position to claim a moral right. In short,
the term "person" is not an ontological but a functional con-
cept; in saying that someone is a moral person we are merely
asserting that he may lay a certain claim upon us and that we
acknowledge a responsibility to or for him. In some degree
children may be persons, lunatics may be persons, animals

may be persons, associations and societies may be persons,
and any there be, may be persons. It is, in my
divinities, if
judgment, one of the fundamental faults of the traditional
liberal philosophy to misconceive the use of the concept of a
person and, quite without reason, either in logic, metaphys-
ics, or morals, to restrict the application of the term person
to individual featherless bipeds.
The second main point to be made is that liberty is not
confined merely to overt actions of a deliberate or voluntary
sort. A person ought to be free to think, to deliberate, and to
choose, as well as to act. Indeed, he ought to be free in any
respect in which it makes sense to speak of him as confined,
hampered, or restricted. Nor does it appear to be at all obvi-

ous that any one such mode of freedom is more fundamental
than any other. It is persons who should be free, and their
liberties are all their own. There are some who would argue
that thought should be free, but only because thought is
for the sake of action. This strikes me as fallacious. For it
assumes that a particular form of freedom requires defense
or justification. Thought should be free to the extent that
persons think, and to the extent that it makes sense to speak
of a person's thoughts as controlled, confined, or hampered. If
a man's thoughts are not subject to control, then his thought
cannot be free, either. But, manifestly, there is a sense in
which what a man thinks can be controlled, and a sense
therefore in which a person can be confined or hampered
through the determination of his thoughts. Again, although
freedom of choice should not be restricted, it is not true, as
Hume and others have argued, that the concept of freedom
begins only with choice and has no application to what lies
behind it. To restrict the freedom of persons to their choices,
especially if one limits the application of the concept of a per-
son to adults, may well be, in many cases, to countenance by
implication perhaps the worst of all forms of oppressors the —
over-zealous guardian or parent, such as the utilitarian, James
Mill. If children are persons, then there is always a prima
facie case against disciplining them in any way. This is not
to say, of course, that discipline is always unjustified; it is to

say that itmust always be justified, whereas the right to be
at liberty need not be.
Thirdly, neither the concept of liberty nor the principle of
liberty has any necessary or even any very close connec-
tion with such other notions as self-interest, private interest,
or individuality. As I have already said, the author of the
essay "On Freedom" wisely pointed out that no sharp
Social
distinction can be drawn in practice between private and
public interests, and he quite properly insisted upon the
fact that most forms of private and self-interested activity
involve social relations of the greatest importance both to the
individual himself and to his fellows. In any case, the prob-
lem of liberty itself does not turn on such questions as these.
To my mind to defend a liberty only on the ground that its
exercise merely of concern to the individual person him-
is
self or on the ground that it serves his own self-development

is to compromise it from the start. Moreover, other-regarding
impulses, I should have thought, have quite as much right to

be unhampered as self-regarding ones. No doubt busybodies
are a nuisance, and no doubt most of us rather like to help
ourselves so far as we are able. And the fact that busybodies
are a nuisance may well be a reason to restrict their other-
regarding activities in certain directions. But the liberty of
a busybody, if he is a person, is as much a liberty as that of
the hermit; as such, worthy of respect as that of any-
it is as
one else who is concerned only with his so-called private
affairs. Again, the liberty of altruistic organizations to con-
duct their activities is not something which needs justifica-
tion on the ground that it conduces to the general welfare or
on the ground that it makes possible a greater freedom of
action for those whom they benefit. Liberty, once more, does
not require to prove its case. What wants proving is the case
for its limitations.
Fourthly and finally, it is no part of my intention to argue
that the principle of liberty should replace the principle of
covering law of moral and
utility as the great social action.
There are no moral laws, and none of them is completely
covering. All persons have a right to be at liberty. But prin-

ciples of justice or fair play, of security, of truth, and of
promise-keeping also impose independent responsibilities
upon us. One fundamental error of Mill's essay is that it
sometimes suggests that the only reason why a liberty of one
person may be restricted or curtailed is to protect the liberty
of another. To my mind, this is absurd. Which of the several
primary moral principles by which we live should take
precedence? I think that no general answer can be made to
such a question. It has always seemed to me that the most
artificial as well as the most uninteresting part of Plato's
moral philosophy is his tendency, in the Philebus and else-

where, to arrange abstract goods or virtues in an absolute
hierarchical order of merit without regard to the particular
circumstances within which any good must be actualized or
any obligation fulfilled. What we ought to do in a particular
situation is not always determinable by mechanical resort
to obvious covering principles which tell us that justice
should be done or liberties defended. In practice, every
moral principle contains within itself an unwritten "unless
clause" which provides a basis for the making of exceptions.
The principle of utility provides one basis, though certainly
not the only one, for making exceptions to particular moral
principles. In that case,however, it cannot be used, as Mill
and the author of the essay "On Social Freedom" use it, as a
supposedly higher-order principle for the justification, as well
as the qualification and correction, of all particular moral
practices. And not conceivably does it provide a defini-
tion, as Stephen Toulmin and others have in effect regarded

it, of what we mean by a genuinely moral practice.
[XV]
George Santayana:
Natural Historian of
Symbolic Forms
santayana's philosophical intentions

The death of George Santayana has not been the occasion for
any searching reappraisal of his work. He is ignored by the
philosophical cognoscenti. Nor are his more serious works
widely read as literature. Why this is so is not difficult to
understand, even when one believes, as I do, that he is one
of the few great philosophers of our age. The times are out
of joint. In the older and larger sense, Santayana remained
a moral philosopher, a seeker after wisdom in a period
whose more influential minds have been preoccupied with
exact logic and the methodology of science. Recalling Berg-
son and Whitehead, Alexander and Heidegger, it is possible
to claim that there has also been a strong undertow of meta-
physical interest, and perhaps a place may be claimed for
Santayana in their company. Yet the comparison is unillumi-
nating, for even as a metaphysician Santayana remains apart
from such as these. The controlling and never forgotten
interest of his metaphysics is always moral. Like the ancient
Stoics and Epicureans, he was animated not by a driving
curiosity concerning the nature of things, but rather by a
search for the underpinnings of the rational life. Fundamen-
tally, his materialism is hardly a theory at all. It was intended
neither as a scientific cosmology, as was the evolutionary

materialism of Herbert Spencer, nor as the necessary out-
come of a logical analysis of the basic categories of being, as
is that of contemporary naturalism. Santayana's doctrine of
substance, for example, is in reality a sort of linguistic ritual
in terms of which he pays homage to the factuality —or
—
mystery of existence. To call it "material" is hardly more,
for him, than to evince one's natural piety before it, to sig-
nify, as it were, that one bows to its royal unconcern for the
Thus conceived, materialism provides no
affairs of the spirit.
further description of the natural world which could con-
ceivably add anything to what the physicist may tell us
about it. Its "truth" is dramatic rather than logical or empiri-
cal. To regard it is to misconceive its function and
otherwise
to reify the verysymbols by which it is expressed.
If Santayana is right, however, all metaphysical systems
are to be read in this way. Perhaps this is not the only way to
view them, but it might well serve to rekindle interest in
metaphysics in quarters where that subject is currently re-
garded as the graveyard of meaningless linguistic confusions.
So construed, it may be argued that Plato, Spinoza, and even
Hegel at least make sense, as unquestionably they do not if
one tries, as some do, to read them as super-scientists who
have their own more esoteric methods for discovering what
there really is. Perhaps, indeed, the "real" itself is at bottom
a term of appraisal, by means of which we serve notice of
what most profoundly concerns us. Thus, though metaphys-
ics may have nothing to say to us as scientists, it may still
say much that is vital to us as men. If so, metaphysicians
will have to regard the poets and the prophets, rather than
the scientists, as their natural allies. Nor can they expect to
interest those victims of "scientism" who are prepared to
consign to the limbo of "emotive meaning" every form of
human utterance that cannot be confirmed by the procedures
of experimental science.
Nearly everyone pays lip service to Santayana's powers as
a stylist. Yet to many, by no means all of them analytical
George Santayana: Natural Historian of Symbolic Forms 317
philosophers, his manner of writing has seemed essentially

unsuitable to its genre. Under the present dispensation,
philosophical discourse is not regarded as a branch of literary
art. And its inclusion by university administrators under
the rubric of the humanities is resented almost as much by
the philosophers themselves as it is by professors of literature.
Its more properly conceived, is said to be the logical
aim,
analysis of concepts and the clarification of propositions. For
such a purpose, evidently, what is wanted is a style like that
of Bertrand Russell, stripped of hyperbole and embellish'
ment, and exhibiting at once to the analytic eye the bony
structure of its argument. Clarity, economy, and objectivity
are its sole standards of perfection; ambiguity, vagueness,
and rhetoric its mortal enemies. Now in the hands of a
Russell such a style, at its best, may possess great elegance
and charm. When employed by a Spinoza its very austerity
and impersonality may also be powerfully expressive of a
noble mind that is able to view all reality under the form of
eternity. But whatever its uses, it is not a style which was
possible to Santayana. On occasion, Santayana could cut like
a master to the heart of a logical confusion, though he does
by a process of discursive thought.
so intuitively rather than
Something of power may be witnessed in such little
this
masterpieces as "Some Meanings of the Word 'Is'/' or in the
devastating "Hypostatic Ethics," which sufficed to convince
Russell himself of the confusions involved in an ethics that
reduces goodness to a simple quality. Yet such efforts were
sporadic, and we have Santayana's own word for it that
dialectics went much against the grain. His more natural
manner is dialectical only in the free, ironical way of Soc-
rates in the earlier dialogues of Plato. Like that of Socrates,
his dialectic always wears the aspect of a half-playful im-
provisation, to be forgotten once its moral point is under-
stood. Santayana's writing is also clogged with imagery and
laden with all the tropes that are so maddening to a literal
mind. It represents, or so it is said, the mind of a poet who
for some perverse reason insists on dealing with subjects
inherently alien to the poetic imagination.
Santayana himself, of course, violently opposed the poetics

implicit in such a view as this. Indeed one of his major
contributions to contemporary critical theory was to dispose
of the myth that there is such a thing as a poetic idiom or
subject-matter. In his own case, at any rate, the style is
surely the icon of amind which was incorrigibly concrete
and an imagination which was incurably pictorial. As I shall
presently try to show, these qualities served Santayana's
philosophical purposes very well, but they have plainly
blocked the way to a better appreciation of his work on the
part of some, at least, who like himself have found in the
study of symbolic forms a powerful clue to the understanding
of man.
Moreover, even among those who still respect the name of
wisdom, Santayana largely remains unhonored. Here, it must
be confessed, the grounds for indifference appear somewhat
stronger. It cannot be denied that, as Russell once put it,
Santayana was a "cold fish." Unlike Dewey, or Russell him-
self for that matter, Santayana was not deeply engaged by
the great issues of our lives. He remained untouched by the
terrible anxieties of his generation, and aloof from its pro-

founder loyalties and betrayals. His own commentary, rich
and racy as it so often is, seems at times to belong rather to
the domain of natural history than to moral philosophy. He
defines with apparently equal relish the lineaments of slaves
and freemen, or of romantic barbarians and rationalistic
liberals, always subtly intimating that all are worshiping false
gods. Yet his basic motivation is always clear. Distraction, on
all levels, remained Santayana's special devil; and on all
levels distraction is the special curse of modern life. In

America, he thought, it reached its final apotheosis when, in
pragmatism, distraction finally achieved the status of an
official philosophy. He found here every spontaneity and
freedom save that for which alone he yearned, the freedom

of the spirit to behold, without attachment or care, the realm
of essence. Santayana's only cure for sick souls is inward re-
moval to the life of contemplation, but for such a cure there
is no political prescription.
What Santayana preaches, in so far as he can be said to

preach anything at all, is a personal ethics for the undis-
tracted individual whom, for the nonce, nature permits to
loll some quiet backwater of existence. Such an ethics can
in
have little meaning save for cultivated men of leisure who
have no serious mortal attachments. What it seeks is pure
vision. In Santayana's case, however, its special objects were
the forms and patterns of human culture. What he most de-
lighted to contemplate were the ideal stages of human prog-
ress, removed from their accidental historical embodiments.
Compared to this, the aestheticism of a Pater or a Berenson
remains meagre and half-hearted. For unlike Santayana, the
latter required the continual stimulus of individual works or
movements of art. But as Santayana himself acknowledges,
works of art did not in the end suffice to satisfy his poetic
imagination. They tied him down to the perspectives and
emphases of the particular artist; they imprisoned his spirit
within the limits of alien forms and compelled his mind to
attend to matters which were not of his own choosing.
Santayana preferred a wider perspective than any poet af-
fords and a freer deployment of his imaginative powers than
any decent work of art could possibly permit. In short, he
wished to be unhampered by all adventitious material
integuments whatever, so that he might follow in his own
way the ideal fulfillment of any individual or collective pur-
pose that happened to take his fancy. What he sought was
the internal rationale underlying any cohesive domination
or power, and when he found it he could usually manage to
call it good. He had a predilection for order and solidity,
and it was this no doubt which caused him to prefer aris-
tocratic hierarchy to democratic confusion and disorder. But
his vision, at least, was catholic. The only lesson here, if such
it be, is that the manifest form of any great collective aim or
destiny, when witnessed sympathetically at a sufficient dis*

tance, may serve to delight an imagination to which nothing
human is finally alien.
Whatever its faults, such an aestheticism as Santayana's
has certain obvious merits. Within limits, it tends to breed
consideration, tolerance, and urbanity. And certainly these

traits are among the more attractive by-products of Santa-
yana's point of view. He is never parochial, never fanatical,
never unfeeling, so long at least as the object of his sympathy
is willing to keep its distance and asserts no claims. The
egoism, if you like, is profound. But it is not cruel or self-
assertive. It is possible to prefer it to the harm which many
better men have done.
It is not, however, my desire here to defend Santayana's
aestheticism as a way of life; my concern will be rather with
the qualities of his defects, and with the products of those
For if the truth be told, although Santayana is often
qualities.
accounted wise, I cannot deeply believe it. In what sense,
finally,can this hardly human detachment be wisdom at all?
Or be such, then it is of no sort of which such harassed
if it
and bedeviled creatures as ourselves can make much use. By

a strange irony, which Santayana himself might have rel-
ished, his own way of life is for us very little more at bottom
than a beguiling pose, the apparent form of an alien purpose
which for the time being we are willing to contemplate
because its possessor expresses himself so well. But at last we
are compelled to say that any wisdom so imperturbably in-
different to the more exigent needs of men is, if not a con-
tradiction in terms, then at any rate the product of a fantas-
tic inversion of human functions, the ultimate proof, if such
were needed, of the absolute plenitude of being.
santayana's contributions to the

study of symbolic forms
Yet the very qualities that repel us,when we view them
made possible a unique and original contribution to
morally,
what may well turn out to be the major intellectual achieve-
ment of our age, the philosophy of symbolic forms. Precisely
because of his hypertrophied disinterestedness and his
strange lack of ordinary human commitments, Santayana
was left free, while still living within our civilization, to
examine its controlling symbols and myths. His attitude
toward them, if not his method, is almost that of a scientist.

There is a profound difference, however, between the equip-
ment required of the natural scientist and that necessary to
the student of symbolic form. In the latter case the principal
data are not external material events observable by all who
have the eyes to see. Here the data are meanings, the point
or significance of which cannot be understood without the
guidance of sympathetic intuition. To grasp the full import
of the idea of Christ in the Gospels, for example, it will not
suffice simply to observe the external behavior of Christian
believers. One must also be able to experience, as only a
Christian may do, the unique suasion of that many-sided
idea. Only so can one become aware of what the data
fully
for the study of the Christian myth Meanings, no
really are.
doubt, are human responses; but if one remains unresponsive
oneself, the character of any meaning will remain obscure.
For successful criticism one must also be a primary inter-
preter.
Because of his peculiar background, Santayana was ex-
posed from early age to many winds of doctrine. And so he
acquired at the outset a sensitivity to their inward signifi-
cance which an outsider can achieve only by long years of
loving and laborious study, if at all. At the same time, how-
ever, circumstances prevented him from ever settling into a
single cultural mold. He spoke English like a native, yet
English was not his native tongue. His familial influences
were mainly Catholic, yet a true believer he could never
become. Thus by second nature he was destined to remain
an ever-sympathetic observer rather than a committed par-
ticipator in the complex cultural life about him. By good
fortune he knew at first hand the suasive powers of the
symbols in terms of which the members of our civilization
express, commend, protect, and order their values. Yet he
remained sufficiently aloof from their spell never to confuse
his own primary interpretations of them with his philosophi-
cal gloss. He could both reify the symbol or its function and
accurately note the fact; he could personify when the myth
required, yet clinically observe the personification; he could
project his emotions at an artist's or a prophet's bidding, and

still know the projection for what it is.
These faculties of second nature would not have served
for Santayana's special purpose, however, without the poet's
special sensitivity to the ways of words. This gift freed him
forever from the temptation to worship paraphrase. Literal-
mindedness has its uses; but it is the bane of those who
would grasp the sense of the life-symbols of art, morality, or
religion. Because he was a poet, Santayana knew intuitively
the volatility of words, their constitutional powers of associa-
tion, their subtle affinities and mutual aversion. Language,
for him, was not and could never be a calculus. He under-
stood, as only the poet can, that vagueness, ambiguity, and
metaphor are not diseases of language, and that in the hid-
den meanings of a word may lie its chief glory.
But even this rich and varied endowment would still have
left Santayana inadequately prepared for his special voca-
tion. This required also a philosopher's special powers of
abstraction, by means of which he might accurately discern
in some particular trope the paradigm of a more general
symbolic form.
In this respect also Santayana was peculiarly fortunate.
For the same power, in another, sometimes blinds him to the
very features of symbols to which Santayana was naturally
so sensitive. Santayana probably had less command of math-
ematics and natural science than any other first-rank philoso-
pher since Hume. His education was overwhelmingly hu-
manistic and literary. In a philosopher with the primary
interests of a Russell or even a Whitehead this incompetence
might well have blocked the way to any lasting constructive
effort. But in Santayana's case it merely freed him once and
for all from that fatal worship of science and mathematics

as the models of perfect communication, which until very
recently has impaired the philosophical study of other di-
mensions of human discourse. The fact that Santayana not
only spoke the common tongue but could speak no other
immunized him, as it were, from the syntactical virus of
descriptivism in all its forms. Unlike Russell, Santayana was
never tempted to regard ordinary language as merely an un-

conscious repository of outmoded metaphysics; and unlike
the logical positivists it never occurred to him that ordinary
language might be no more than an obsolescent progenitor
of some ideal language of science. Thus he did not need to
relearn the lesson, as so many philosophers have had to do,
that the special qualities of an ideal language, if such a thing
could be constructed at all, would radically unfit it for every
other occasion for which men use symbols. As it was, he saw
very early that the languages of poetry, religion, and morals
may possess a distinct theoretical interest not only for the
anthropologist or clinical psychologist, but also for the phi-
losopher of symbolic forms. The net result was that Santa-
yana was equipped as perhaps no one else in our time has
been for the philosophic study of the spiritual symbols of
western culture. In this respect only Cassirer, who possessed
a profounder erudition but lacked Santayana's sensitivity, can
be compared with him.
HIS INTERPRETATION OF RATIONALITY

Perhaps the most distinctive feature of Santayana's portrayal
of the cultural symbols of art, morality, and religion is his
placement of them within the life of reason without at the
same time forcing them into the molds which alone count as
rational in formal logic or empirical science. There are, he
contends, standards of adequacy and relevance, and hence
of reasonableness, for the criticism of the former which in no
way depend upon the assumption that works of art are
logical systems or that religious myths are scientific hypoth-\
eses. Rationality is a wider concept than logical deducibility
]
or scientific verifiability. Its core meaning involves the no- ^

tion of order; as applied to human activities in any sphere, it\
primarily connotes fitnes s, propriet y, and-adjustafc ility. Wher- (
ever there are proprieties, there also, at least implicitly, are ?
standards of reasonableness and unreasonableness. Given /

them, we can provide a rationale for judgment in any do-
main; without them judgment lapses into an expression of
arbitrary preference or animal faith. This means, of course,

that any rational art or morality must be traditional, for with-
out tradition standards of propriety would have no meaning.
An art that was individual talent would be an art to which
all
rational criticism would be irrelevant, just as a world of ex-

perience in which every item were unique would be a world
in which the principles of rational understanding could gain
no purchase.
Rationality, then, is the regulation and coordination of
activity by ordering principles or rules. As in the case of
natural languages, these may be and implicit.
largely tacit
But where they operate, there is nevertheless a usage which
defines what it means to be rational and, hence, irrational.
Irrationalism in the domain of knowledge implies precisely
the setting up of individual judgment or intuition against
established canons of right thinking and evidence; in religion
it means a rampant protestantism which has no standard of
integrity beyond the intensity of personal feeling and

commitment; in art it means a restless romanticism which
places novelty above style, and holds originality in self-
expression to be an end itself; and in politics it means con-
tempt for law, treaty, and diplomacy. In each of these
spheres Santayana himself was a traditionalist, but what
this meant to him fundamentally was devotion to the life of
reason. If he fears democracy, then this is because he views
democracy as Socrates doubtless viewed it at his trial, as the
epitome of lawlessness and unreason. If his customary ur-
banity is temporarily in abeyance when he is criticizing
Fichte or Hegel, then this is because he regards their philos-
ophies as the embodiment of an unbridled wilfulness inflated
to cosmic propositions. Hegel may call his system rational,
but in the misuse of that term he is, for Santayana, commit-
ting a double crime against reason.
However, Santayana's preference for the life of reason
does not prevent him from seeing the importance of under-
standing and sympathizing with irrationality. Here we come
upon another distinctive feature of his analysis. I have said
that Santayana approaches the study of symbols somewhat
in the spirit of a scientist. In this connection it would be more

illuminating to compare him to some logician who found the
study of fallacies and illogic as absorbing and, in its way, as
illuminating as the formulation of the principles of valid
inference. In logic or in science such an interestwould have
no point. The logician sometimes desultorily catalogues a few
types of fallacy, but only in order to put us on our guard
against their commission. As such, fallacies have little im-
portance for logical theory. When an inference has been
shown to be invalid, the logician forthwith dismisses it.
Similarly once a scientific hypothesis has been shown to be
false, it retains interest only for the historian or pathologist
of human thought. But in the case of other symbolic forms
such indifference to the invalid and irrational is not always
justified. Just as ambiguity and vagueness have their uses
for the poet if not for the scientist or logician, so in morals or
religion irrationality may have a function which is vital to
the ethical or spiritual development of mankind. If they did
nothing else, the irrationalists would still hold before us a
kind of inverted mirror in which we may more truly see our
parochial and custom-ridden selves. By going outside the
rules they force us to re-evaluate our habitual adherence
to them and so to decide for ourselves whether, as it
stands, may be worth while to continue playing the moral

it
game. Thus they may also compel us to see the need to

modify the rules in order to make room for new values. In
this way, indirectly, they may even strengthen a moral
order which has become inflexible and hence is weakened
by every vital interest which cannot find a place within it.
As Santayana shows us, in effect, the life of reason itself has
its uses for irrationality which reasonable men do well to
acknowledge. For only when we occasionally look beyond

the limits of reason can we reasonably decide whether its
boundaries have been well drawn.
In this way Santayana absolves himself of any charge of
rationalistic formalism. Discipline is as essential to rational
art or rational religion as it is to rational morality. But all of
these activities are more than disciplines, and they serve
needs which are more vital than any interest in order for its
own sake can ever be. We play games in accordance with

rules, — —
but unless we are mad not for the rules' own sake.
Within art, morality, or religion there is always a post-rational
aspect which reasserts itself for the sake of delight, or pas-
sion, or ecstasy. Yet even the post-rational itself may gradu-
ally be rationalized as reason extends its bounds in order to
become more hospitable and more humane. Thus, most
broadly conceived, the ideal of reason itself may be envis-
aged, without fatuity, as a gentle, encompassing order which
is coterminous with joy. Since no need would ever arise to
be aware of them, such an order would know no bounds, but

they would still be there, doing their silent work of civilizing
our passions and domesticating the urgencies that we call
will.
No doubt such a vision of the ideal of reason is fanciful.

But or something like it, nevertheless keeps the mundane
it,
life of reason open at its further end, so that the stream of
post-rational aspirations and values may gradually filter

through. In this way we learn to preserve our sanity while
at the same time continually enlarging its scope. Madness is
always the counterpart of a rigid rationality faced with a

world beyond itself which it cannot comprehend. Stated in
another way, every healthy and adequate symbolism must
have its symbols of self-transcendence, boundlessness, and
inexpressibility. By learning thus to say what cannot be said,
we manage by the art of paradox to reduce the margin of
terror and to give sanity even to madness itself. There are
no questions, finally, which have no answers. This is the
final commitment of the life of reason.
On its more ordinary planes, as Santayana conceives it,
the life of reason is any life which, at the respective levels

of common sense, science, art, morality, or religion knows
how to guide and correct its attitudes and beliefs in the light
of certain ordering principles. These in turn reflect, at a dis-
tance, the respective communal demands in response to
which they themselves were instituted. In any of these do-
mains, criticism is saved from being a merely ad hoc symp-
torn of first-personal preference or prejudice by the invoca-

tion, as it were, of a social rite, which the
in relation to
individual critic functions as a spokesman and mediator.
Rational criticism requires the giving of reasons, and reason-
giving presupposes relevant procedures in accordance with
which, in each case, we make the moves of a solemn game.
Only in thisway do we ever become aware of the meaning,
within any sphere of human life, of the distinctions between
appearance and reality, or between irrationality and ration-
ality.
REASON IN CRITICISM
It is with The Sense of Beauty that aesthetics may be said to
have come of age in America. That work has already become
a classic; it is doubtful, indeed, whether its leading ideas
have since been greatly clarified or improved upon by the
aestheticians. Its limitations are serious, but it is Santayana
himself who in his own later writings shows the way to a
more adequate philosophy of art. The Sense of Beauty is
not, plainly, a work by which its author sets much store. For
it still moves within a traditional framework of artificial con-
cepts which he subsequently found to have generated false

issues completely remote from the interests of art or criti-
cism. In the end the whole subject of aesthetics became
distasteful to him, and its problems came to be viewed as
the irrelvant consequence of a series of accidents in the
history of ideas. Later, whenever he had occasion to speak
of it or its practitioners, Santayana did so condescendingly
and disparagingly.
It is essential to our present purpose to inquire into the
philosophical causes of this disenchantment. In this way we

may somewhat better comprehend with what serious intent,
in The Life of Reason, Santayana wrote of reason in art, or
why, in Three Philosophical Poets, he regarded such a work
as De Rerum Natura as a great poem and not merely as a
derivative philosophical treatise that happened to be written
in verse. When the reasons behind these views are brought
to light, it becomes startlingly clear how profound was Santa-

yana's opposition to prevailing philosophical conceptions of
the so-called aesthetic attitude. What he really proposed, in
effect, was a revolution in critical theory, the full significance
of which we only just now begin to comprehend. For by in-
sisting, both as a theorist and as a practicing critic, upon the
relevance of the full symbolic content of a work of art to our
appreciation and appraisal of it, he was virtually maintaining
the irrelevance of so-called aesthetic analysis and judgment.
How, then, did it happen that after what seemed such an
auspicious beginning, in a subject for which he was appar-
ently so eminently qualified, he thenceforth resolutely turned
"his back upon aesthetics? One clue may be found in the
essay "What is Aesthetics?" which is included in the volume
Obiter Scripta. What he
there describes is not an organic
discipline at but a factitious medley of history, psychol-
all,
ogy, morals, and bad philosophy. It is a subject which al-

together lacks any unifying concept that could bring into
clearer focus the full range of what we hold intrinsically
valuable in art. The term "aesthetic" itself, at any rate as it
is currently understood, is pre-eminently unsuitable for such

a unifying purpose, especially as applied to the symbolic arts
of poetry and painting. It actually diverts attention away
from nearly everything in a work of art which gives it sig-
nificance and value. Taken seriously, it forces us to regard
the complex beauty of The Divine Comedy as hardly more
than a witless swill of pleasant sensation, or the myriadic
levels of significance in Hamlet as sheer unaesthetic irrele-
vance. How could such a concept provide the basis for
serious interpretation and criticism?
Santayana's attitude on this point is easily misunderstood.
Itmust not be forgotten that in his case, unlike that of so
many philosophers and critics in the recent past, this dis-
paragement of the aesthetic was in no way attended by a
corresponding revulsion against contemplative values. On the
contrary, he reacted, as we have seen, ever more intensely
against the pragmatic spirit of the age in which he lived.
Indeed, viewed in one way, his entire later work is a con-
tinuing protest against the distracted obsession of the age

with the instrumentalities of social and political organization
and against its theoretical and moral glorification of activity.
Realms of Being may turn out yet to be Santayana's
masterpiece; but it can scarcely fail to appear anachronistic,
both in attitude and in doctrine, in a period of ascendent

pragmatism and operationalism. What do the activistic psy-
chologies and theories of knowledge engendered by this
mentality have to do with such notions as essence and con-
templation? If the former are taken seriously, then the latter
become correspondingly obscure or meaningless. In Realms
of Being a whole province is inhabited by nothing save
essences, and in The Realm of Spirit, possibly the most af- *
fecting of all his works, Santayana devotes his finest prose to \
the delineation and praise of the contemplative life. He did /
not thereby renounce empiricism and naturalism; but in )
Realms of Being what we find are an empiricism and a\
naturalism very remote from the prevailing philosophies^
which share these names.
Because of this proliferation of "realms" in his later work,
some of Santayana's critics have suggested that his ontology
acutely suffers from a problem of overpopulation. This is
mistaken, as a glance at the above-mentioned essay, "Some
Meanings of the Word 'Is'," may suffice to indicate. To ex-
plain this would require a full chapter in a comprehensive
study of Santayana's philosophy of language. Be this as it
may, it is notable that nowhere in all his realms of being does
the notion of the aesthetic find any place or purchase. For
sensation and for sensations there lie ready at hand more
appropriate and less misleading terms; so also for aware-
ness, attention,and pleasure, the other concepts frequently
associated with the aesthetic. As now used, this misbe-
gotten term is intolerably ambiguous, and its use results
merely in theoretical and critical confusion. On the one
hand it appears to refer to the intrinsic interest or satisfaction
we may take in the contemplation of any object; on the
other to the apprehension of what Prall calls "sensuous sur-
face" and Northrop the "aesthetic continuum." Mixed to-
gether, as is usually the case, these completely different

notions yield the useless theory of aesthetic experience as
a delightful but thoughtless arrest of attention in sensation.
What conceivably could be the use of such a concept either
for our understanding of the arts or for our more normal
apprehension of them? The one it empties of all pregnancy
and expression; upon the other it saddles a misplaced notion
of the epistemological "given" that was conceived originally
for an utterly different theoretical purpose. It does not even
provide a helpful semi-technical gloss upon the normal mean-
ings of "beautiful." For when, beneath its sensuous charm
(which there no intention here of disparaging), we discern

is
in a work of art some intimation of its fitnessJ:o a human pu r-

p ose, o r some larger sens e of order and design, or some
poignant articulation of the life-values of a culture, or,
finally, some intensification of our awareness of what it means

to be human, then we do not hesitate to credit that work
with a richness or quality of beauty beyond that usually
ascribed to the sensuous surfaces of images. It is this many-
leveled beauty of the symbolic forms of art which creates
his problem for the critic and calls forth his special resources
as an interpreter. When, as in our own age, a sometimes
"penitent art," as Santayana calls it, renounces all meaning
and representational form in the name of something called
"aesthetic purity," the result, however interesting it may be
in itself, is an art which has gratuitously forsaken its heritage
for the sake of a confused slogan whose very meaning can
scarcely be rendered intelligible. Santayana's protest here is
not against the sensuous, but against a theory of aesthetics

which has nothing to do with our normal transactions with
works of art. Works of art are symbolic forms, and only when
they are so conceived can we begin to grasp the richness of
their moving appeal or their place among the high intrinsic
values of our lives.
In maintaining that not merely does the literary dress or
the incidental play of imagery in Lucretius' poem, but also
its imaginative philosophic content, belong inherently to it
as poetry, Santayana's profounder intention, of course, is to

meaning
restore to the life of contemplation the full range of
and significance from which our modern empiricist theories
of knowledge have tended to divest it. It has become a
dogma of latter-day empiricism, in short, that contemplation
— or acquaintance, as James and Russell used to call it —can
have no object save what is "given" in sense perception.
However useful this doctrine may have been in directing at-
tention to the crucial role of observation in the enterprise of
knowledge, it had also the incidental effect of reducing the
contemplative values traditionally associated with art, as well
as with philosophy and religion, to the level of sensation
and feeling. It thereby renounced exclusively to science the
whole domain of thought and meaning. The result was a
tremendous cultural calamity. The impressive thing about
Santayana is that he could reasonably claim membership in
the empirical tradition in philosophy while at the same time
seeking to restore to the true contemplative interest in art all
of the levels of meaning which most other empiricist aes-
theticians in our time have denied it.
At such a point it is easy to go too far and claim for the
poet a primarily cognitive intention. Santayana, however,
does not do this. Indeed, it is precisely to free the philosophy
of symbolic forms from its preoccupation with the role of
symbols in the communication of knowledge and truth that
he offered his elaborate interpretations of western poetry and
religion. The "idea" of Christ in the Gospels, as Santayana
portrays it,is not a vehicle of knowledge in the ordinary
sense. Its primary function is quite different. So also in the

case of poetry: the poet's basic intention is expressive and
imaginative. Ideas are used in poetry, but they do not exist
there for the sake of information. "The poet's art," he tells
us, "is to a great extent the art of intensifyingemotions by

assembling the scattered objects that naturally arouse them."
In most cases these objects are perceptions or their con-
ceptual counterparts and surrogates. But for the poet their
role is to provide a correlative object in which our emotions
and feelings may be projected; through its perception alone
are they focused and intensified. To accomplish this result
the poet uses, or is free to use, every symbolic resource of

which nature and convention can avail. The difference be-
tween the poet and the scientist is not so much a difference
in the sorts of symbols which each employs; it lies rather in
the different uses and kinds of total response made by each
to their symbols.
Just as Santayana finds himself unembarrassed in ascrib-
ing conceptual meaning to art, so also he is not fazed by
speaking of art as "rational." In a lesser philosopher, this
would doubtless an attempt to "reduce" the arts to
signalize
science or to logic and hence to impose upon criticism a
rationale fundamentally alien to it. In Santayana's case, how-
ever, this is precisely what was not intended. His contention,
at bottom, is merely that great art exhibits certain charac-
teristic features of order, coherence, and unity which alone
entitle us to call it "rational" in the generic sense of that term.
From our awareness of these characteristics we gradually ac-
quire implicit standards of artistic relevance and cohesion
which we then apply in judging works of art and,when
challenged, in the justification of our judgments. The unfit-
ness or unrelatedness which we discern and condemn in
inferior works is not a matter of logical incoherence or in-
validity; the unity of a work of art is a unity of style, feeling,
and mood. But why, if they are the products of orderly ad-
justment, and if, at the same time, they communicate to a
manifold of impressions, thoughts, and images a real sense
of interrelatedness and relevance, should works of art be
denied the title to rationality?
The symbols of great poetry, then, reveal to us a world
unified, orderly, and serene. For the time being at least, this
world is no less objective and no less real than those of ra-
tional common sense or science. Indeed, the very fact that
poetry reveals a world to us suffices to guarantee its rational-
ity.For what we mean by the rational, fundamentally, is by
definition to be distinguished from chaos, confusion, and ir-
relevance. Great art is never a phantasmagoria. Even the
Walpurgisnacht of Goethe, for example, is in no sense a
manifestation of disorder. On the contrary, it is a powerfully
wrought creation of a profoundly ordered poetic imagina-

tion. As such it is a product of rational art. In the older and
deeper sense of the term, every great poem is a design. But
just as the older theologians sought in the argument from
design to demonstrate the overarching rationality of the uni-
verse itself, so the critic in pointing out the design of a work
of art is thereby testifying to the rationality of a system of
symbolic forms. Logic presents us with one type of order
among symbols and is rational because it does so. The arts,
at their best, present us with another type of order among
symbols, and this too may be regarded as rational, although
in a different sense.
Here, curiously, may be discerned an echo of old Kant
who perhaps alone among the classical philosophers of art
understood how the beautiful may imbue us with a sense of
purposiveness and objectivity, and hence of rationality, with-
out at the same time transforming the object contemplated
into an instrumentality of action. It is an echo, I submit,
which deserves to be heard again and again in the sphere of
art criticism and interpretation.
REASON, MORALITY, AND TRUTH
Now have already observed, Santayana's whole phi-

as I
losophy is, moral philosophy. Perhaps
in the broadest sense, a
this explains why he never bothered to write an independent
treatise on ethics. However, there may also be another rea-
son for this omission, not unlike the reasons which lay be-
hind his increasing contempt for the academic discipline of
"aesthetics." In the course of a very funny reply to Professor
Stephen Pepper's essay on "Santayana's Theory of Value,"
Santayana makes the following interesting statement:
. . . Pepper finds my "theory of value" ambiguous. It is so am-

biguous that, under that name, I was not aware that I had one.
I certainly have a doctrine concerning the good, borrowed from
Socrates and his school: a doctrine rather than a theory, since
it professes to be a judgment rather than the description of a
fact. And it concerns the good, the object of desire, rather than
"value." Value is an economic and secondary term like "use."
Things have or acquire value in different connections from dif-

ferent points of view; and a universal history of ethics and eco-
nomics would no doubt contain a theory or description of values,
which might be abstracted from the total picture. But I confess
that when Pepper goes on to distinguish various literal theories
of value, I hardly know what he is talking about. 1
Some critics have inferred from such remarks as these that

Santayana was wholly interested in substantive ethical ques-
tions and that he was indifferent to the linguistic and logical
problems of so-called meta-ethics. But they are mistaken, and,
as I shall show in a moment, it is not difficult to piece to-
gether the outlines of a fairly coherent and always interest-
ing theory of moral discourse from various chapters in his
works from The Life of Reason through Winds of Doctrine
to The Realm of Truth. The fact is rather that, given San-
tayana's own conception of moral judgments and their justi-
fications, he could not but think that whole enterprise of

value theory, as conceived by such writers as R. B. Perry
and Pepper, is fundamentally misdirected. For them the only
interest in answering questions about the meaning of such

terms as "value" and "good" is to provide descriptive defini-
tions of them which will provide a comprehensive conceptual
scheme for a general science or theory of value. From San-
tayana's point of view, such a program mistakenly assumes
that so-called ethical expressions are terms of objective em-
pirical reference and that all value judgments, including
moral judgments, are, with a littleand hauling, logi-
pulling
cally reducible to ordinary statements of fact. For him this
mistake is fundamental. How, then, could anyone speak of
his "theory of value" when he denied the basic premise un-
derlying such a theory? And how could any just critic point
out ambiguities in his theory of value when indeed no such
theory existed in which ambiguities might be found?
But these questions, after all, are a trifle disingenuous. For
"Apologia Pro Mente Sua," The Philosophy of George Santayana,

1
ed.by Paul Schilpp; Vol. II in The Library of Living Philosophers

(Menasha, Wis.: George Banta Publishing Company, 1940), p. 577.
at the beginning of The Sense of Beauty, for example, San-

tayana himself boldly asserts that "the philosophy of beauty
is a theory of values," just as he also says that "A definition
that should really define must be nothing less than the ex-
position of the origin, place, and elements of beauty as an
object of human experience." 2
He also says other things in
that work which may give the impression, not unnaturally,
that he actually meant to equate the meaning of "the good"
or "the desirable" simply with "the desired" as such. Thus he
remarks, "We may
therefore at once assert this axiom, im-
portant for moral philosophy and fatal to certain stubborn
all
incoherences of thought, that there is no value apart from

some appreciation of it, and no good apart from some pref-
erence of it before its absence or its opposite. In apprecia-
tion, in preference, lie the root and essence of all excellence.
Or, as Spinoza clearly expresses it, we desire nothing because
3
it is good, but it is good only because we desire it." Of
course it is ominous, particularly in a writer so
just a little
inexact as Santayana, that as late as the Apologia, in the very
passage which I quoted above (pp. 333-4), he could speak
quite comfortably of "the good, the object of desire" in the
same breath that he emphatically distinguishes a "doctrine"
or "judgment" concerning the good from "the description of
a fact." But it is also true that in The Sense of Beauty, in the
paragraph immediately following that in which the reference
to Spinoza is made, he goes on to say, "It is true that in the
absence of an instinctive reaction we can still apply these
epithets by an appeal to usage. We may agree that an action
is bad or a building good, because we recognize in them a
character which we have learned to designate by that ad-

jective; but unless there is in us some trace of passionate rep-
robation or of sensible delight, there is no moral or aes-
thetic judgment. It is all a question of propriety of speech,
and of the empty titles of things. The verbal and mechanical
proposition, that passes for judgment of worth, is the great
2 The Sense of Beauty (New York: Charles Scribner's Sons, 1936),

p. 13.
3 Ibid., p. 16.
cloak of ineptitude in these matters. . . . Verbal judgments

are often useful instruments of thought, but it is not by them
that worth can ultimately be determined." 4 Whatever else

this fascinating and prophetic passage may seem to imply, it
surely does not suggest that judgments of value are con-
ceived by Santayana as ordinary empirical statements of fact.
On the contrary. Although, as he says, we may agree that an
action is bad because we recognize in it some character
which we have learned to designate by that adjective, this
does not amount to a moral judgment on our part unless our
agreement involves an expression of our own positive ap-
proval. When our own approval is not involved our agree-
ment amounts to a "verbal and mechanical" proposition in
which, so far as we are concerned, the word "bad," con-
ceived as a moral term, is merely idling. Spinoza himself
said that we good because we desire them; but it

call things
does not follow that in so calling them we are merely re-
porting our belief that they are, for ourselves or others, ob-
jects of desire.
It seems reasonable, then, to say that even as early as The
Sense of Beauty Santayana never meant to lend support to

any program, such as Perry's, which would reduce all judg-
ments of value, including moral judgments, to ordinary em-
pirical statements about objects of interest, their causes, ef-
fects, and he later preferred
interrelations. Interests, or, as
to say, "passions," are no doubt the "source" or "seat" of
values. It is entirely with them and the actions issuing from
them that moral judgments are concerned; it is in order to ex-
press or to incite them that such judgments are made; and it
is in order to reinforce them that moral justifications are
provided. In view of their expressive and incitive function, it

is entirely natural that moral judgments should also serve to
inform our listeners or readers of our sentiments and feelings

about things, just as facial expressions and gestures similarly
inform them. In conventional situations, moreover, such
words as "good" and "right" may also be used, in a deriva-
tive, secondary sense, to inform others that the objects to
4 Ibid.,
pp. 16-17. Italics mine.
which they are applied conform to certain conventional

standards in which we ourselves may have no great personal
stake. Thus the butcher tells us that a certain cut is a very
good steak, though he himself detests steak and would, if he
could afford to do so, commend it to no one at any price.
Likewise the art dealer tells us that a certain picture is a
very fine Matisse though he is bored silly by Matisse and sells
his paintings only as a convenience to his customers. Yet, if I
correctly interpret Santayana's view of the matter, such fac-
tual information as these remarks may convey is not what is
involved when we employ such words as "good" in making
genuine moral or aesthetic judgments. Moreover, even when
we assert conventional "propositions" about good beefsteaks
and Matisse, a residual element of commendation remains.
For the idea of a good steak would not connote a steak which
is, say, tender, juicy, and well-aged unless most steak fanciers
would be prepared to judge a steak having these characteris-

tics as something worth eating. In fact, it is precisely this logi-
cal dependence of what Professor Pepper himself calls

"standard value" upon the use of evaluative terms in first-
personal expressions of interest or satisfaction which in part
lies behind Santayana's own claim, in The Sense of Beauty,
that "all values are in one sense aesthetic." For "aesthetic"

judgments are by definition merely expressions of immediate
appreciation which articulate our sentiments and feelings,
without reference to or thought of the conformity of their
objects to any conventional standards whatever. In them,
such terms as "good" are used in their logically most primi-
tive and fundamental sense simply to express our liking or
favor of any object which we directly contemplate, imagine,
or envisage. And it is only because our likings are, for the
most part, fairly stable, and hence our expressions of them
tolerably uniform, that "good" and "valuable" gradually be-
come associated in particular contexts with certain percep-
tual characteristics of which our value judgments are pres-
ently taken as signs.
Such, it seems to me, is a not-implausible reading of San-
tayana's so-called "theory of values." However, it might be
argued that, with the benefit of a good deal of hindsight, I

am reading into Santayana's words in The Sense of Beauty,
just what I am disposed to find there. A more crucial work
in this connection, therefore, is the essay on "Hypostatic
Ethics," which is included in Winds of Doctrine, his devas-
tating critique of Bertrand Russell's early philosophy. It is an
interesting piece, even if, as Santayana remarks somewhat
ironically, his logic in it is "not very accurate or subtle."
Most of Santayana's critics, including Russell and Pepper, re-
member it chiefly for its attack upon the thesis, which Russell
took over from G. E. Moore, that the term "good" refers to a
unique, unanalyzable, and absolute quality whose presence in
any object can be apprehended only by a special form of
non-empirical, ethical intuition. What they forget is that San-
tayana completely agreed with Moore and Russell, as against
the ethical "naturalists," that goodness as such is indefinable
and that it is always a mistake to equate it with any charac-
teristic whatever which may be common to the things men
call "good." It is therefore a nice question whether what
Russell admitted to having learned from Santayana's essay
is precisely what Santayana meant to teach. Let us see.
Santayana's statement of his position is so succinct and so
forceful that it would be foolish of me to try to paraphrase it:
Before proceeding to the expression of concrete ideals, he

[Russell] thinks it necessary to ask a preliminary and quite
abstract question, to which his essay is chiefly devoted; namely,
what is the right definition of the predicate "good," which we
hope apply in the sequel to such a variety of things? And he
to
answers at once: The predicate "good" is indefinable. This
answer he shows to be unavoidable, and so evidently unavoid-
able that we might perhaps have been absolved from asking
the question; for, as he says, the so-called definition of "good"
A—-that it is pleasure, the desired, and so forth—=-are not defin i-
/ Hn n c nf *hp prprlipfltp. "goon
" but designations of the thingsJto
1
I which this predicate is applied by different persons. Pleasure,

ând its rivals, are not synonyms for the abstract quality "good,"
but names for classes of concrete facts that are supposed to
possess that quality. From this correct, if somewhat trifling,
observation, however, Mr. Russell, like Mr. Moore before him,
evokes a portentous dogma. Not being able to define good, he

hypostasises it. "Good and bad," he says, "are qualities which
belong to objects independently of our opinions, just as much

as round and square do; and when two people differ as to
whether a thing is good, only one of them can be right, though
it may be very hard to know which is right." "We cannot main-
tain that for me a thing ought to exist on its own account, while
for you it ought not; that would merely mean that one of us is
mistaken, since in fact everything either ought to exist, or ought
not." Thus we are asked to believe that good attaches to things
for no reason or cause, and according to no principles of dis-
tribution; that it must be found there by a sort of receptive
exploration in each separate case; in other words, that it is an
absolute, not a relative thing, a primary and not a secondary
quality. 5
At this point, unfortunately, Santayana's talk of qualities

leads us a bit off the track. He makes it appear momentarily,
that the fundamental issue between him and Russell is only
whether goodness is an absolute and primary quality which
is predicated of things "as they are in themselves," or a rela-
tive and secondary quality which is predicated of them only
under certain standard, operative conditions. Were this the
real issue between them, then it would be hardly necessary
to take sides, since the whole distinction between primary
and secondary qualities remains notoriously unclear, and
since it may plausibly be argued that when any term, rela-
tive or absolute, is literally predicated of a subject it really
and absolutely is predicated of it, just as it is also true that
the use of all predicative terms, whether the "qualities" they
represent are secondary or primary, presupposes what may be
called "a context of predication." Thus, when we say that
something is "heavy" or "sweet" or "good," we are talking
about the object, not about our view of it, our sensations of
it, or our feelings toward it. And the same is equally true
when we say, a trifle more complexly, that something is to

the right of something else or even that it causes certain re-
5 George Santayana, Winds of Doctrine (London: J.

M. Dent &
Sons, 1940), pp. 140-1.
actions in us. But it is also obvious that we would not say

sincerely that heavy if, for example, we do not think it so,
it is
if it simply disappears from view whenever we try to weigh
it, or if the situation in which we make the statement is not
one in which it is understood that we are speaking of things

of comparable weight. If one is talking about animals it is
true, absolutely and unconditionally, that elephants are
heavy. But when the talk is about planets, it is not merely
false to say that elephants are heavy, but grotesque, ridicu-
lous, or even senseless. But such remarks as these take us
farther and farther from the real point at issue between Rus-
sell and Santayana. For it is no part of Santayana's intention
to deny that when we judge something to be good we really

mean to be saying something about it; nor is it any part of his
purpose to assert that when we judge something to be evil
we mean to say merely that it appears evil to us, though per-
haps not to someone else. In sum, I take it that it is not his
view that nothing is good or evil but thinking makes it so. 6
Santayana's own relativism is not in the least Protagorean; it
turns on a different point altogether. But again it is well to

let him speak for himself:
. . . what suggests this [Russell's] hypostasis of good is rather

the fact that what others find good, or what have we ourselves
found good in moods with which we retain no sympathy, is
sometimes pronounced by us to be bad; and far from inferring
from this diversity of experience that the present good, like the
others, corresponds to a particular attitude or interest of ours,
and is dependent upon it, Mr. Russell and Mr. Moore infer
instead that the presence of the good must be independent of
all interests, attitudes, and opinions. ... To protect the be-
lated innocence of this state of mind, Mr. Russell, so far as I
can see, has only one argument, and one analogy. The argu-
ment is that"if this were not the case, we could not reason with
a man as to what is right." "We do in fact hold that when one

man approves of a certain act, while another disapproves, one
of them is mistaken, which would not be the case with a mere
6 I myself doubt very much that anyone at all in his right mind ever
meant to maintain precisely this view.
emotion. If one man likes oysters and another dislikes them, we

do not say that either of them is mistaken." In other words, we
are to maintain our prejudices, however absurd, lest it should
become unnecessary to quarrel about them! Truly the debating
society has its idols, no less than the cave and the theatre. The
analogy that comes to buttress somewhat this singular argu-
ment is the analogy between ethical propriety and physical
or logical truth. An ethical proposition may be correct or incor-
rect, in a sense justifying argument, when it touches what is
good as a means, that is, when it is not intrinsically ethical, but
deals with causes and effects, or with matters of fact or neces-
sity. But to speak of the truth of an ultimate good would be a
false collocation of terms; an ultimate good is chosen, found, or

aimed at; it is not opined. The ultimate intuitions on which
ethics rests are not debatable, for they are not opinions we
hazard but preferences we feel; and it can be neither correct nor
incorrect to feel them. 7
Now, making due allowances for the fact that Santayana's

idioms are not those of many contemporary analytical phi-
losophers, I think it is plain that the fundamental gist of these
remarks is precisely that moral judgments are logically of a
sort completely different from statements of fact, whether
"natural" or "non-natural"; that they are made in order to ex-
press commitments or preferences, not in order to state be-
liefs and that this being
or opinions about questions of fact;
argue about their "truth" or "correctness,"
so, it is pointless to
at least when we have passed beyond the point where a sin-

cere consensus of moral sentiments or attitudes may be pre-
supposed. If, past this point, we persist in saying to those
with whom we disagree in attitude that our own judgments
are "true,"we will be doing nothing more than misleadingly
reaffirming our loyalty to them. And when we admit, despite
our present convictions, that they may be "false" or "incor-
still
rect," this is way of indicating our own

hardly more than a
willingness to reconsider them if and when it can be shown
that in continuing to hold them we would be false to our
own true heart of hearts. In a word, to admit the possibility
that a "fundamental" moral judgment may be false is only a
T Ibid.,
pp. 143-4-
way of suggesting that it may turn out to be not quite so fun-

damental or unalterable after all. Making such admissions to
those with whom we morally disagree is merely a way of
holding out an olive branch, to preserve the peace until we
have had time to search out, and perhaps to find, a more ul-
timate area of agreement within which our present moral
differences may gradually be resolved.
Thus, long before the current vogue of so-called "non-cog-
nitive" ethics, Santayana was already saying in substance
that morality, conceived as a form of first-personal judgment,
is not and can not logically be reduced to a science of any
sort, and hence that what we can plausibly mean by such
phrases as "rational morality" and "moral truth" must be

something quite different from what many philosophers have
had in mind in speaking of "scientific ethics" and "the truth
about morality." This reading of Santayana becomes indispu-
table in the intriguing chapter on "Moral Truth" in The Realm
of Truth. In that work, he grants that the moral command-
8
ment, "Love they neighbor as thyself," "purely hortatory as

this seems may be almost entirely translated into propo-
. . .
sitions that would be either true or false." 9 For embedded in

the commandment is the knowledge that all living things are
ends to themselves as much as we to our own selves. Yet, as
he says, this truth "is not, and cannot become, a moral com-
mandment," since "The categorical nerve of every imperative
is vital, it expresses an actual movement of the will." * For
both Santayana and Kant every principle of practical reason

must contain within itself what Kant calls a "determination
of the will," and every moral principle is essentially a practi-
cal principle which, as such, can never be translated into any
8 The Realm of Truth, in my opinion, is one of Santayana's best and

most neglected works.
9 Realms of Being, one- vol. ed. (New York: Charles Scribner's Sons,
1942), p. 475.
1 Op. p. 476. In this instance, Santayana's idiom, interestingly
cit.,
enough, very close to that of Kant, which, incidentally, shows how

is
far, despite his formalism and his rationalism in ethics, the great "China-
man of Konigsberg" really was from what passes nowadays for "cog-
nitivism" in ethics.
purely speculative theory about what purport to be matters

of fact.
Santayana also makes a great point of the distinction be-
tween morality proper, which is concerned with the articula-
tion of our "actual allegiance in sentiment and action to this
or that ideal of life," and the "descriptive science of ethics,"
as he calls it, which describes "the history of such allegiances,
and of the circumstances and effects involved." 2 In fact, it is
this distinction, which lies at the root of so many of Santa-
yana's most penetrating, yet always sympathetic, criticisms of
the philosophies and theologies of his predecessors. For him,
it is the blurring of this distinction which is responsible in
large part for the fabulous moral histories of the Jews and
Christians, the hypostatic, moralistic metaphysics of the Pla-
tonists, the mythic philosophies of history of the Hegelians
and the Marxists, and the transcendental philosophies or
"sciences" of mind to which so many German philosophers
have been addicted since the time of Kant himself. The same
confusion is also responsible for the various ad hoc epistemol-
ogies that philosophers construct out of whole cloth in order
to provide at least a verbal account of the meaning and veri-
fiability of propositions whose ambiguous functions are thus
systematically run together. As a detached man of the world,
Santayana is rarely censorious of the forms of life that are
embedded in these doctrines, and so long as they are sin-
cerely avowed he is prepared to respect them for the vital
human concerns which they may unconsciously express.
What he deplores, as a lover of truth and wisdom, are the
comprehensive which they give rise, the false com-
illusions to
fort which they afford to their victims, and the terrible way in
which they help to harden natural moral affinities into meta-
physical dogmas and theological creeds. Himself an instinc-
tive traditionalist and aristocrat, Sanatayana is always im-
mensely respectful of Plato as a philosophical moralist. He
objects only to Plato's tendency to convert his own moral
perspectives, by a piece of logical legerdemain, into a mytho-
logical cosmology whose peculiar "truth," quite understand-
2 Ibid.,
p. 473.
ably, apprehendable only by a specially trained philosophi-

is
cal elect, whose unique powers of insight extend beyond the

understanding of ordinary scientists and mathematicians and
the lay moral intuitions or sentiments of the ruck of mankind.
Santayana's spiritual Catholicism is always put on the stretch
when he discusses anything German. But even German ego-
ism he can understand when conceived as a sincere preoccu-
pation with the affairs of the self. His objection is that the
German idealists, mistaking the nature of the informal moral
dialectic of the soul's discourse with itself, converted it into a
monstrous organon or method for the construction of elabo-
rate Geistesunssenschaften which purport to tell us, at one
and the same time, both what there is and what there ought
to be. The always chaster English logicians, although pas-
sionately insisting on the logical distinction between what is
and what ought to be, convert what ought to be into a queer
realm of being populated by strange, unanalyzable qualities,
whose relations it becomes the province of an unnatural sci-
ence of ethics to describe. And once more, in order that man
may come to know the principles of that science, there is the
tell-tale postulation of a special faculty of ethical intuition
which, independently of all human attitudes and preferences,
is supposed to give us that truth about morals which is also
the moral truth.
For Santayana it was thus the part of wisdom to know our
moral principles for what they are, the all too human ex-
pressions of organic passions and sentiments. So conceiving
them, we are no longer tempted to press beyond the congeni-
tal "naturalism" of undeceived common sense in order to
postulate moral half-substances which would matter less to
us if we knew them for what they really are. When we know
what we are up against, we are left free candidly to avow
and to defend the way of life embodied in those precepts we
ourselves find most congenial, without dogmatically opposing
others whose principles are affirmed on grounds that are
logically as impeccable as our own. Nor will we then seek that
spurious assurance of all metaphysical moralists who suppose
that the unrighteous would be automatically converted if
only they were required to take a proper course in moral

physics.
Here, however, we must make another turn of the linguistic
screw. In his later writings, as he perceivesmore clearly the
fundamental logical point of between morality
his distinction
and the descriptive science of ethics, Santayana is no longer
so reluctant to admit that there may be such a thing as moral
truth. No longer anxious lest such an admission may seem
to countenance a mythical moral physics, he is left free to
acknowledge the familiar fact that ordinary men also some-
times claim truth for their moral principles. It is now only a
question of explaining correctly what such claims amount to.
His suggestion is that although fools, in morals as elsewhere,

are always ready to cry, "How true!" to any proposition
which is widely held, expressing thereby little more than
their own diffused and cowardly wish to belong to the crowd
and to share in its sentiments, men of integrity will reserve
the title of "moral truth" for considered judgments which
they sincerely and steadily profess in the light of the best
knowledge of themselves which moral intuition or insight
may afford. However, once again he tirelessly reaffirms the
point that "This Socratic self-knowledge is not scientific but
expressive, not ethical but moral; and here if anywhere, in
the discovery of what one ultimately wants and ultimately
loves, moral truth might be found." 3 And then he at once so
wisely adds, "This is no easy discovery; and we must be pre-
pared for surprises in morals, no less than in physics, as in-
vestigation and analysis proceed. As the blue vault vanishes
under the telescope, so moral conventions might dissolve in
an enlightened conscience, and we might be abashed to per-
ceive how disconcerting, how revolutionary, how ascetic the
inmost oracle of the heart would prove, if onlywe had ears
to hear it. Perhaps a premonition of this ultimate moral disil-
lusion rendered Socrates so endlessly patient, diffident, and

ironical, so impossible to corrupt and so impossible to de-
ceive." 4
3 Ibid., 480.
p.
* Ibid.
The basic error of Socrates, as it is also the error of all dog-

matists, is the assumption that human
nature is single and
immutable and that "the good that glimmered like buried
gold in his own heart must lie also in the hearts of others,
and only ignorance or sophistry could keep them from seeing
it."
5
It is for this reason that Santayana, who opposed any at-
tempt to stretch moral unity "beyond the range of natural

organization," considered it the part of wisdom, as well as of
logic, to recognize the ultimate diversity of morals. And al-
though "politically, and within the living organism . . .
moral dogmatism is morality itself," it remained for him spirit-

ually "a sinister thing, a sin against the spirit elsewhere."
Where, we may ask, is elsewhere? To this, I suppose that
Santayana would reply, that region, secret to every man, in
which, like Kierkegaard, he "suspends the ethical" in order
that he may achieve a vision of that quiet and serenity which
is spiritual liberation. This does not mean that the liberated
spirit may not return to the forum or to the battlefield of the
moral life where, like everyone else, he is bound at last to
speak for a moral truth which he cannot sincerely disown. It
means, rather, that he returns in effect as one twice-born, who
views the necessity of returning at once tragically and ironi-
cally, as a condition which befits a being created in the image
of God yet compelled, whether he will or not, to dwell forever
in the realm of matter.
There is a final, closely related, error of the sons of Socrates
which the philosopher Santayana would remove. Here, curi-
ously enough, he once more comes close to the pragmatists,
particularly to William James. James, it will be remembered,
spoke of "rationality," the love of order, coherence, and har-
mony, as a "sentiment." And it was his claim that when this
sentiment wars within our breasts for regimen over other
passions it must make its own way and prove itself by its own
fruits. It has no antecedent metaphysical claim to priority,
and if we ourselves say that it ought to prevail, why then we

are once more expressing our own submission to it in the face
of other claims which we thereby find less exigent. But we
s Ibid., p. 481.
may of course deceive ourselves, and there is no natural

harmony, written into either nature or the logos, which pre-
cludes the possibility that we may discover later that at our
heart's core, "whirl is king." Similarly, while proclaiming
throughout a long life the merits of the life of reason against
barbarians and protestants and romantics of all sorts, Santa-
yana also insisted that the truth about morals everywhere does
not include the thesis that no principle may count as "moral"
which cannot be universalized or which defies the moral prin-
ciples of concord and consistency. What, when all is said and
done, if we find that we prefer discord? And what if our
moral passion for equality or freedom refuses to give way, in
cases of conflict, to the sentiment of rationality? In that case,
by a curious paradox, does not that sentiment itself then be-
come unreasonable in demanding an organization, whether
personal or interpersonal, to which our unruly natures will
not submit? And do we not then subvert the name of moral
truth in proclaiming its universal validity, when we know
that we cannot avow it sincerely and with an undivided
heart?
Some have misunderstood Santayana at this point. Of all
men, he perhaps the last to preach irrationalism. What he

is
does preach is the cause of self-knowledge and the refusal of

true moralists to say, as a certain self-styled Christian platon-
ist and was once overheard to say in a moment of
socratist
apoplexy, "Jesus is reason!" Reason is a principle or complex
of principles, among the most precious, certainly, which a

moral philosopher may avow. But it is not, properly, an ob-
ject of worship, not a surrogate-God, before which all men
must bow their heads and bend their knees. Santayana may
have worshipped no god, and, as some will think, that was
his final limitation or trouble. His great merit as a philosopher
consisted in the fact, that at any rate, he worshipped no false
gods, including the philosophical god of reason. And this, in
my opinion, helped him clear his head for the study of the
treacherous symbolic forms which men call "morality."
CONCLUSION
It is my belief, then, that Santayana is best understood as a
natural historian and philosopher of symbolic forms. So con-
ceived, his work takes on a relevance for both contemporary
philosophical analysis and critical theory which has a signifi-
cance that this brief study has barely intimated. Still I would
not claim too much. When
one presses Santayana beyond a
certain point for answers, they are often not forthcoming. At
the crucial point he is more likely than not to put one off
with a flashing metaphor or figure of speech, when what one
wants is definition and plain talk. One feels that he is more at
home in the middle region between general theory and spe-
cific interpretation or judgment. Indeed, this is why in my
title I have preferred to call him a natural historian rather
than a philosopher of symbolic forms. It is hardly worthwhile,
however, to dwell upon his limitations, which are well known
to all, when his contributions to the philosophy of meaning
are so little known and so poorly understood. He has given us
insights, as I have tried to show in the case of aesthetics and
ethics, which we have hardly as yet begun to recover. If
there were space, much the same could be shown to be true
of his theories of religious discourse. If he lacked fundamen-
tal warmth, this very fact made possible the detachment
which was necessary for his probing studies of the controlling
symbols of western culture. If on occasion we are outraged by
his fantastic indifference, we may at least be grateful for its
fruits. Man, it has been said, is the talking animal. Santayana,

through a long life, contributed much to our understanding
and therefore to our knowledge of
of the nature of his talk,
the animal himself. No more can be claimed for any contem-
porary philosopher.
\% vl y* yi yt yt y* y* y* y* y* v*
[XVI]
Philosophical Analysis
and the Spiritual Life
A REPLY TO MR. WILLIAM EARLe's

"notes ON THE DEATH of culture"
MR. EAKLES CRITIQUE OF MODERN CULTURE

Academic philosophers, by the nature of their calling, are
menders and preservers. For them, existence is a subject for
discussion, and language something to be mentioned but
never used. You would never guess from their immaculate
publications that doomsday may be in the offing. Western
civilization may be cracking up, but no signs of the fact are
to be found on the pages of The Journal of Philosophy or
Mind. There, if nowhere else, the Logos remains undisturbed;
there anxiety is acknowledged only as a concept, and con-
sciousness exists, if at all, only as a phenomenon, like dust,
black-spot, or the great bend of the Orinoco. Professor Wil-
liam Earle, that is to say, is something new under the aca-
demic sun, at least in America; and his disturbing essay,
"Notes on the Death of Culture," x would deserve notice if for
no reason but the fact that he not only loathes his "subject"
but says so in so many words. What is still more impressive,
however, is his readiness to generalize; the dreadful predica-
ment of contemporary professional philosophy is, for him,
merely a symptom of the dry-rot which is destroying our en-
tire Western culture.
1 Noonday I. New York: The Noonday Press, 1958.
As Mr. Earle himself points out, his theme is not new; for
over a century unacademic philosophers, historians, and men
of letters have debated the causes of what Matthew Arnold
called "this strange disease of modern life." More original and
intriguing are Earle's diagnosis of the ailment, and his "pri-
vate" dream of a new, authentic culture which might con-
ceivably arise from the ashes of our existing pseudo-culture.
For him, there is no hope of reviving what pundits call "The
Great Tradition." He admires it, somewhat nostalgically, for
what it was, but he believes that it is now without a saving
remnant, and he boldly calls therefore for a complete and
revolutionary break with what has gone before. Like Marx,
and unlike Nietzsche and Kierkegaard, Earle conceives of
this break as a social and not merely as an individual neces-
sity. But like Nietzsche and Kierkegaard, and unlike Marx,
he conceives it in purely spiritual rather than in material

terms. What is wanted, in short, is a completely new vision
of human destiny in which the activities of religion, art, and
philosophy may again be fused, as perhaps they were in
biblical times, in one great communal act of cultural crea-
tion.
If Mr. Earle's intentions are revolutionary, there is some-
thing slightly archaic, at least, about his diction. In our time
the concept of culture has been largely appropriated by bac-
teriologistsand social scientists. But Mr. Earle makes it clear
from the outset that he is not talking about phenomena of
any sort, whether microscopic or macroscopic, whether in-
dividual or social; he conceives of culture in a purely ideal
sense, as "the whole life of the human spirit in communities."
This sounds a bit like Hegel; as we shall see, the idea behind
it is also Hegelian. In Earle's philosophy, as in Hegel's, it is
the concept of spirit itself which is the operative term. What

is spirit? Earle's answer is somewhat muffled, but it provides,
nonetheless, the essential clue to his diagnosis of our cultural
predicament. It must be understood, in the first place, that
the life never "automatic or instinctual": "it
of the spirit is
must be created by the spirit itself." This means, among

other things, that the spirit must be concerned, concerned
Philosophical Analysis and the Spiritual Life 351
"for itself, for its life" as a "free effort." It means also a striving
toward self-consciousness, toward lucidity about what it it-
self is and what other things are. And it means, finally, that
the spirit must manifest itself to others in works lucidly ex-
pressive of its self-concern. Such a conception invites parody.
I myself am reminded of a line in one of W. H. Auden's
poems about "the high, thin, rare, continuous worship of the

self-absorbed." The proper analogy, however, is again with
Hegel, whose talk about spirit as something which exists in,
for, and through itself is plainly echoed in Earle's very turns
of phrase. There are some, I know, who will take this com-
parison itself as an implied rebuke. It is not so intended. And
if the complaint is seriously pressed that Earle has utterly
failed to define the essence of spirit, the only and sufficient
reply is that spirit, like existence, has no essence. Indeed, the
very attempt to define "spirit," like the attempts to define
"existence" or "good," is the root of many philosophical evils.
The uses of such expressions may be explained, but that is
another matter. Earle does not attempt this, nor, I suppose,

would he consider it a useful philosophical venture.
But there are many ways of clarifying an idea, and Mr.
Earle has at any rate managed to specify three characteristic
and inalienable "functions" of spirit, as he calls them. And it
is precisely in their dissociation that he locates the spiritual
derangement of modern culture. In our time, unfortunately,

"religion" no longer has anything to do with matters of creed
or commitment; like Royce's loyalty to loyalty, it has become,
on the contrary, a sheer concern with concern itself. What
passes for "philosophy" is no longer a love of wisdom or an
attempt to clarify a total vision of human destiny, but an ob-
session with lucidity as an end in itself. And what we call
"the arts" are no longer expressions of human emotion or
representations of recognizable forms with which men in real
life are deeply concerned; they are, rather, vacuous efforts at
"pure" expression or meaningless arrangements of "aesthetic

surfaces." In a word, "religion," "philosophy," and "the arts,"
no longer serve as names for dimensions of an existentially
undivided spirit; they have become mere disciplinary titles
for sterile academic subjects and professional activities which

have lost any vestigial sense of their common spiritual vo-
cation. Meanwhile philosophers analyze, theologians dread,
poets howl, painters arrange pieces of wrapping paper, mu-
sicians make sounds, and sculptors bend hairpins. All are
quite mad, and like madmen, all are spiritually dead. They
have nothing to do with culture, and culture has nothing to
do with them.
Mr. Earle's diatribe also reminds one of Wyndham Lewis.
Unlike Lewis, however, Earle has no personal animus. He is
talking more of trends than of the achievements of particular
men. He does not make the mistake of denying the special-
ized talents of the Picassos and Schonbergs, the Russells and
Wittgensteins, the Bubers and Tillichs. They are sometimes
clever or amusing; they may, for a time, titillate the senses or
intrigue the intellect; but in the end, as Mr. Earle sadly con-
fesses, they always become a bore. And boredom, of course,
is the final and most terrible enemy of spirit.
THE STATE OF PHILOSOPHY IN

THE AGE OF ANALYSIS
As it stands, such a diagnosis is excessively abstract. It would
be improper therefore to criticize it until we have first seen
how it is fleshed out in Mr. Earle's more detailed criticisms of
what is now happening in philosophy, religion, and the arts.
His description of the condition of contemporary philosophy,
naturally enough, is the fullest and best informed, and it is to
this that I will give closest scrutiny.
Symptomatic of the decline of philosophical culture in our
age, according to Earle, is the emergence of the "technical
philosopher," a fabulous creature whose sole professional in-

terest in the clarification of ideas has lost all contact with the
spiritual ends which alone give point to the demand for
lucidity. Earle portrays him, with great gusto, as an intellec-
tual eunuch, obsessed with narrow "analytical" questions of
logic, meaning, and method which, by definition, have noth-
ing to do with reality and, because of this, are devoid of con-
tent. This frivolous logic-chopper, whose clever little word-

games are as immaculate, impersonal, and witless as the
computations of a Mark III, is then compared, somewhat to
his disadvantage, to thatparagon of serious intent, the "tradi-

tional" philosopher, a man of large and "edifying" visions who
has somehow managed to transcend the ordinary man's un-
philosophical concern with such particular evils as sickness,
poverty, war, and tyranny in order to meditate without dis-
traction upon the sheer meaning of existence.
Earle names few names, but one of his examples of the
analytical school is J.
S. Mill, whom Nietzsche contemptu-
ously referred to as a "blockhead" and whom an admirer
called, perhaps not inconsistently with Nietzsche, "the Saint
of Rationalism." Earle's classification of Mill is not unjust but
misleading. Mill did write a famous treatise on logic and sci-
entific method in which he had a great deal to say about the

problem of meaning. But, as in his other works, his concern in
writing it was philosophical. That is to say, it was part of
his life-long campaign against the dogmatism and apriorism
which stand in the way of human enlightenment and prog-
ress. Mill hated the so-called Reason of the intuitionists and
transcendentalists because it short-circuits criticism and forti-

fies entrenched opinion and habit under the guise of insight.
And he respected logic and science because he believed that

in them lies our only hope of destroying the idols which have
enslaved men's minds since the dawn com-
of history. Earle's
plaint against Mill, however, is that he supposed that the
suffering of human life may be analyzed into a plurality of
specific correctable ills. What Mill failed to see, according to
Earle, is that the suffering of life is a problem which "requires
for its solution an inner philosophic transformation of our atti-
tude toward life." To this last, however, the only reply is that
Earle has evidently forgotten Mill's moving Autobiography
or such pieces as his fine appreciative essay on the arch-
transcendentalist Coleridge. It is true that Mill thought
deeply and often effectively about specific remediable miser-
ies; it is not true that he had no conception of the "pathos of
human existence," or the necessity of facing it with philo-
sophic fortitude and understanding. Quite the contrary, utili-
tarianism, in itsmost extended sense, was for Mill a way of

life, as it had been also for Bentham. And if it partially failed
in its Benthamite formulation to satisfy all of his spiritual

needs, Mill continued to revise and to enlarge it throughout
his career.
In Mr. Earle's jaundiced eyes, more recent analysts lack
even Mill's diversified interest in human life. They spend
their time tracking down metaphysical absurdities which
they ascribe exclusively to the misuse of ordinary language.
Or else they devote themselves to the construction of un-
intelligible systems of symbolic logic or to the creation of im-
possible "ideal languages" of science from which every trace
of spiritual meaning has been relentlessly eliminated. In
brief, the contemporary analyst of whatever school is a fanat-
ical methodolatrist and logomaniac who no longer remem-
bers, if he ever knew, what existence really means. Bereft of
vision and devoid of style, his vacuous little papers on the
meaning of meaning are so tedious that not even his own
friends can keep awake while they are being read.
What shall we say to all this? I believe that, to begin with,
we must acknowledge, with chagrin, its grains of truth. Most
professional philosophers, let it be said, are not large-scale
dreamers, dedicated, at least in their waking hours, to the
pursuit of wisdom and the edification of their fellows; they
are, in aword, clods. And this is as true of the professional
metaphysicians as it is of the analysts. At our best universities
there are numbers of men now writing and teaching some-
thing called "philosophy" who would find the craft of the
greengrocer taxing and the simple integrity of a virtuous
chimney-sweep beyond their powers to emulate. They not
only have no vision; they have no ideas. And their inflated
academic reputations as "scholars" depend entirely upon
their familiarity with the lore and jargon of their profession
itself. Such men, I agree, are not practitioners of philosophy,
but its scholastic spawn.

So much, and more, needed to be said, and we may be
grateful to Mr. Earle for saying it so entertainingly. But
when since the morning of the world have things been differ-
ent? Plato's heaviest ironies were directed at sophists who

teach nothing for a price. And the Bishop of Cloyne —him-
self, as it happens, a first-rate analyst and a profound student
of the philosophy of language —directed all his efforts against
those "minute philosophers" who first words
raise a dust of
and then complain that they cannot see. No century has ever
produced more than a bare handful of authentic philos-
ophers, and in every century the Spinozas, the Kants, the
Mills, and the Russells have had to fight their way against a
ruck of academic deadbeats who are always on hand to
—
throw the book of tradition or culture at anyone who —
manages to think for himself. Now it is easy, I know, to exag-
gerate, just as it is also easy to undervalue, the achievements
of one's immediate predecessors and contemporaries. But
even if we limit the list to primarily analytical philosophers,
it would be hard deny the term "authentic" to such origi-
to
nal thinkers as Russell and Wittgenstein. And by extending
somewhat the notion of analysis, we might also include the
American pragmatists, Peirce, James, and Dewey. In various
ways, all of these thinkers have thought profoundly about the
nature of language, thought, and truth. It could be argued,
indeed, that they have jointly effected a revolution in our
very conception of the life of reason and hence of culture.
And most of them have written with as much distinction and
style as the philosophers of any age.
Merely to say this, however, is to play Mr. Earle's own
game, and I, at least, decline to play it. For, despite Earle,
Morton White's phrase, "The Age of Analysis," is not exact as
a characterization of the twentieth- century in philosophy; nor
does White himself pretend that it is. Such systematic think-
ers as Whitehead and Santayana were men of vision in pre-

cisely Earle's own sense; they sought through philosophical
study and contemplation to find a vision of reality which
would provide the basis of a truly philosophic way of life.
But unlike Earle they were not hostile to the analysis of con-
cepts, and both of them used their own analytical powers to
great effect in clarifying the fundamental concepts and cate-
goriesby which contemporary men must try to live. I am also

puzzled by Mr. Earle's reluctance to mention the
just a little
phenomenologists and existentialists with whom in the past
his own work has been most closely identified. It cannot be
that he regards the existentialists as analytical philosophers,
for their anti-intellectualism and anti-rationalism is as great
as their manner of writing is inexact. True, it could be argued
that the phenomenologists, in their own odd way, are analysts
who believe that by a special form of inspection or intuition
they can discern certain primordial forms of human experi-
ence. And Husserl, one of the founders of the phenomeno-
logical school, was certainly interested in problems of logic.
But, as White remarks in The Age of Analysis, Husserl be-
lieved that his own analytical method could also shed light
on art, religion, law, history, "and all other aspects of culture
and the universe." In short, ours is by no means exclusively
an age of analysis; nor are the analysts as devoid of concern
for the life of the spirit as Earle pretends.
But with the exceptions of Santayana and Whitehead,
there is one fact about contemporary philosophy of all schools
that distinguishes it from the philosophy of earlier times. This

is its inability to construct or, in many cases, its lack of inter-
grand style of
est in constructing philosophical systems in the
a Spinoza or a Hegel. And for this no apology need be made.
What remains illuminating in Spinoza's philosophy is not his
rationalistic metaphysics and theology, but his acute and
highly technical analyses of the passions of the soul and the
sources, in them, of human bondage. Likewise what remains
of interest in the ponderous tomes of Hegel is not his deadly
ontologic, which Kierkegaard parodied with such devastating
effect, but his immensely suggestive side-glances at human
institutions and his vivid sense of the mutability of all things

human, including the life of reason itself. Much the same, I
should judge, will turn out to be true of the philosophies of
Whitehead and Santayana. If one thing is dead in contem-
porary philosophy, it is system.
But there are reasons for this which Mr. Earle fails to con-
sider. In fact, the fundamental weakness of his description of

the present philosophical scene is precisely its lack of that
sense of dialectical historical development of which Hegel
was so acutely aware. Earle wholly ignores the fact that there
are compelling historical reasons why philosophers in our age
distrust the very idea of system. He also ignores the shifts in
perspective which have disposed his contemporaries to ques-
tion even the intelligibility of those edifying visions which he
ascribes, with something less than complete accuracy, to the
traditional philosopher. Historically, the gnostic's vision of
went together with a characteristic theory of knowl-
reality
edge and a metaphysics which seemed to provide its philo-
sophical foundation. This was as true of Plotinus as of Spin-
oza or Leibniz. To most philosophers since Kant, however,
such a theory of knowledge and such a metaphysics are no
longer credible. In fact, the whole preoccupation of contem-
porary philosophers with problems of logical and linguistic
analysis is an inevitable consequence of the running critique
of traditional ideas of reasonand reality initiated by Hume
and Kant. Certainly the end is not clear. But anyone who
dreams to any purpose of the future of philosophy cannot dis-
miss a critique of reason which, for over two centuries, has
transformed the very notion of the philosophical enterprise
itself.
In his own way, Mr. Earle himself recognizes the futility of

trying to go back again to the visions of his predecessors.
What he must be asked, therefore, is whether his own gnostic
conception of wisdom, which is so deeply entangled with tra-
ditional rationalistic notions of human knowledge, is any
longer viable, even for those who, like himself, reject the
view that philosophy consists merely in the logical analysis of
concepts. Like many others, Earle believes that the tradition
of Western philosophy has come to a dead end. Yet at the
same time he yearns toward an ideal of the philosophical
quest which lies at the heart of the tradition which he rejects.
Now I hope that I understand something of Mr. Earle's fun-
damental ambivalence; like Nietzsche, I also share it. But we
must cut our losses and accept the fact that there is no longer
a Logos at the heart of things. Thenceforth we must go it on
our own.
PHILOSOPHICAL VISION AND

THE NATURE OF LANGUAGE
Why are contemporary philosophers unable to believe in the
idea of a Logos and hence in the ideal of a philosophical
vision of things which has any noetic significance? The fun-
damental reason concerns a traditional philosophical concep-
tion of language and hence of thought itself which is just now
in its death throes. In order to make this clear, it will be
necessary to say something about that conception itself; only
then will we be in a position to estimate the contribution
which the analytical study of language can make to the un-
derstanding and the preservation of our common culture.
It should be emphasized at the outset that, contrary to the
impression conveyed in Earle's "Notes," technicality is noth-

ing new to philosophy. Indeed, from Plato to Spinoza most
traditional philosophers believed that true philosophical vi-
sion is itself attainable only through the study of exact sci-
ence, mathematics, and formal logic. Plato himself assigns to
these disciplines a higher place than poetry and music on the
hierarchy of cultural activities precisely because the forms
which are their objects are closer to the form or idea of the
Good. In short, like all true gnostics, Plato assumed that the
fundamental significance of all cultural acts is noetic, and
hence that the spiritual order of merit is at bottom a cognitive
order. For him, even goodness itself is conceived mainly as an
object of knowledge, and man's own highest good consists in
its sheer apprehension. This doctrine, however, is rooted in
preconceptions concerning the nature of thought and, ulti-
mately, of language which have been fully exposed only in

very recent times.
Here it is necessary to correct another impression con-
veyed in Mr. Earle's "Notes." He intimates, if he does not
actually state, that the interest of technical philosophers in
—
the nature of language is something peculiar to our own age.

Actually, the philosophical preoccupation with language is at
least as old as Plato himself. From his Cratylus to Berkeley's
astonishingly prophetic Alciphron, the problematical rela-
tions of language to thought and to the "objects" of thought
—
"what there is" have been frequently, if also inconclusively,
debated by philosophers of all types and schools. In fact, it
has frequently been argued by Bertrand Russell and others
that such philosophers as Aristotle took far too seriously the
syntax, parts of speech, and terms of ordinary language as
essential clues to our understanding of the categories of being
and even of being itself. In my own opinion, they were, in
one sense, right to do so; that is to say, it may be that our
knowledge of "what there is" will turn out to be, or at least
to depend upon, a clear understanding of what we are doing
when we say that something exists, that something is good,
that there is a God, that there is a number between one and
three, or that space is infinitely divisible. But this is plainly
very different from what the traditional philosophers seem
to have had in mind. In their case, as also in the case of a
great many later philosophers including Russell and the early
Wittgenstein, it was assumed, in effect, that when language is
used meaningfully it provides a kind of relief map of that
portion of reality which it is said to represent or signify. And
it is largely for this reason that Russell, like Leibniz, regarded
the study of the logical grammar or syntax of language as

philosophically important. For him, the philosophical analy-
sis of language matters because the grammar of a truly clear
language would disclose the underlying metaphysical struc-

ture of reality.
This traditional conception of language is closely related to
the also common view, most succinctly expressed by Hobbes,
that "words are wise men's counters." According to this no-
tion, as it has been developed in more recent times, any
proper or healthy language is a kind of calculus whose "trans-
formation rules," as they are called, prescribe the exact logi-
accordance with which alone any significant
cal conditions in
complex statement may be translated (or reduced) into con-
junctions of logically basic or simple propositions. The latter

propositions, in turn, are held to be meaningful only if their
semantical elements or terms each signify, in virtue of an
explict "formation rule," some distinct and perspicuous item
of experience. Superficially at least, such a theory may seem
reasonable enough. But it plainly implies that ordinary lan-
guage, which has no such rules, and which contains many
terms, such as "spirit," whose "objects," if they exist at all,
are exceedingly obscure, is necessarily an imperfect vehicle
for precise thought or communication. It implies also that
such forms of utterance as poetry, which admits of no precise
translations or paraphrases, or morals, whose essential terms,
such as "right" and "ought," literally signify nothing at all,
must be consigned to the semantical ash-can of "pseudo-
statements" and "pseudo-concepts," as Rudolf Carnap calls
them.
Whether culture, in Mr. Earle's sense, can survive such
rough-shod treatment without perversion is very doubtful.
And if this is what positivism in its contemporary forms comes
to, then I, likehim, want none of it. The important point, how-
ever, is that it is not his traditional philosophers who provide
effective tools for fighting such a rampant scientism but
rather the so-called philosophers of "ordinary language,"
whose preoccupation with the ordinary ways of words
Earle, like Bertrand Russell, so deplores. In fact, the tradi-
tional philosopher could not effectively oppose positivism,
since positivism merely the historical outcome of his
is itself
own basic misconceptions concerning the nature and func-

tion of meaningful discourse. He too was profoundly afflicted
with the same philosophical virus of "descriptivism" which
leads those who suffer from it to assume that all significant
utterances are mere descriptions or signs of what exists in
"the real world."
Actually, however, the real philosophical tradition is far
more complex than either Mr. Earle or I, up to this point, has
suggested. Already in the eighteenth century, "the old happy
time" as Nietzsche nostalgically called it, a new conception of
language was emerging which in the end was to sound the
death knell to the whole rationalistic and gnostic tradition.

The first philosopher to oppose the traditional view was the
Irishman, George Berkeley. Berkeley, himself a spiritualist
and churchman, saw at once that the traditional religions of
the West could not survive, were the view ultimately to pre-
vail that meaningful "trains of words" should be treated
merely as signs of "trains of ideas" ( or perceptions ) which in
their turn signify the true order of things in the external
world. Even if such a view would do, as it will not, as an
account of the language of science, maims every other
it cul-
tural activity that depends essentially upon the use of lan-
guage. Accordingly, Berkeley argued in effect that, not only
in religion and ethics, but also in mathematics and logic,
words do not normally function as "signs" at all. Their mean-
ing, he contended, can never be determined by following a
priori that prime rule of all "literalists of the imagination,"
"Hunt the referent!" On the contrary, it can only be found by
examining in the contexts of their ordinary uses what we, as
speakers, do with them.
But it was not only in his conception of language that Berke-
ley thus anticipated the later thought of Ludwig Wittgen-
stein, the prime mover of contemporary linguistic philosophy.
Berkeley's philosophy of language was conceived primarily
as a weapon for the defense of the spiritual life against
would-be atheists and materialists. In short, like Wittgenstein
and his followers, Berkeley used linguistic analysis "therapeu-
tically." That is to say, he sought by its means to iron out the
strange "mental quirks," as Gilbert Ryle calls them, that have

landed traditional metaphysicians on philosophical "Queer
Street." Much the same is also true of the celebrated "critical
philosophy" of Kant, whom Bertrand Russell once spoke of,
contemptuously, as "the greatest mistake in the history of
philosophy."A large part of Kant's purpose in his Critique of
Pure Reason was precisely to expose the basic "category mis-
were responsible in the past for such philosophical
takes" that
pseudo-sciences as "rational ontology," "rational cosmology,"
"rational theology," and "rational psychology." At every point
Kant's own essentially therapeutic criticism of these mis-
directed "disciplines" amounts to an exposure of the philo-

sophical misues and misapplications of such terms as "cause"
and "substance" and "existence" to "things in themselves." He
argued, accordingly, that there are no real answers to the
questions of traditional speculative metaphysics, but only the
philosophical cure which occurswhen we see why such ques-
tionshave no real meaning. For Kant, the true interest in
metaphysics is moral and religious. And for this reason, he
treated the "transcendent" ideas of freedom, immortality,
and God as principles or "postulates" of "practical reason."
Kant's successors, the great German idealists, continued his
attack upon the traditional tendency to subordinate the prac-
tical to the speculative use of reason, and hence, to misapply
its standards of meaning and truth to the so-called "objects"
of religious belief, moral endeavor, and artistic expression.
Properly understood, a large part of their work was directed
to freeing what Hegel called the "symbolic forms" of Geistes-
wlssenschaft from their historic subservience to ideals of
lucidity and knowledge that properly hold only in the sphere
of Naturwissenschaft. In one way, indeed, Hegel's much-
ridiculed dialectic or "logic of being," which he also called,
significantly, a "logic of passion," is at bottom only a mis-
guided effort to disengage the spiritual activities of art, reli-
gion, and philosophy from the stifling hold, not of science
itself, but of the perennial scientism and logicism of the tradi-
tion. But Hegel forgot the precautionary lessons of his mas-
ter, and once again the forms of 'logic" were treated as a
virtual map of the forms of being. So far in this direction did
Hegel go, in fact, that for him the dialectic itself became a
law of historical development, and the developing ideals or
aspirations of the human came to be hypostatized,
spirit
paradoxically, as successive "moments" in the historical self-

revelation of that obscure substance he called "the absolute."
In short, Hegel once more reduced all affairs of the spirit to
a special form of spiritual knowledge or "science," whose
higher value depended for him entirely upon his own con-
ception of its underlying cognitive significance. Thus, as
Kierkegaard perceived, Hegel remained in the end a gnostic
in spite of himself. And his extreme "panlogism," as it has

been called, is perhaps the most extreme example of what
a rampant gnosticism can come to. At bottom, however,
Hegel's obscure efforts to explain and defend the life of the
spirit against the philistines went awry precisely because of
his faltering grasp of the drift of the symbolic forms through
which that life is articulated.
It is a commonplace that contemporary analytical philoso-
phy arose, in the first instance, primarily as a reaction against
the self-discrediting metaphysical paradoxes of idealism and
its outlandish claims regarding the nature not only of human
history but of the universe as a whole. Whatever their faults
and limitations, such philosophers as G. E. Moore and Ber-
trand Russell attacked idealism primarily in the name of
lucidity; what the idealists said was confused and even unin-
telligible. If this is philosophy, then philosophy is an offense
against the intellect and therefore against the human spirit
itself. The irony is that Russell's fundamental weakness is of

the same sort: that is, his contempt as a logician and a phi-
losopher for the ordinary uses of words. But his attempt to
establish a truly "scientific" philosophy at any rate rested on
an expert knowledge of logical and scientific procedures. Be-
cause of this, he did not fall into the trap which proved the
undoing of idealism: the representation of art, morality, and
religion as vehicles of a higher form of cognition which, by
obeying the laws of a peculiar spiritual logic, could give us a
knowledge of reality inaccessible to the ordinary scientific
understanding. And his celebrated 'logical atomism," so far
from being a wanton effort to destroy the underlying unity of
the spiritual life, was an heroic, if also misdirected, effort to
hold it together. It represented merely his hope of discover-
ing those simple, perspicuous forms of thought which, as he
believed, represent the same basic ontological patterns which
are everywhere embedded in our experience of what there
is. Like Plato, Russell believed also that through the study of
logic and the "ontology" which it reveals, the philosopher
may attain to that contemplative understanding of reality to
which rationalists and gnostics of all ages have aspired.
The fact is that Bertrand Russell, the arch-exponent of

mathematical logic and conceptual analysis, is merely the lat-
est, and possibly the last, great exemplar of the rationalistic-
gnostic tradition which Mr. Earle so much admires and whose

decline he so much deplores. Russell is still looking for that
logical needle in the ontological haystack, the word which
will reveal the Word behind the "word." For him "knowl-
edge by description" is at best a poor sign and necessary sub-
stitute — —
for time-bound men for that direct acquaintance or
intuition which provides a vision of what is really there. Rus-
sell admits, to be sure, that men do other things with words
than formulate the knowledge about reality with which, un-

happily, finite minds must make do. But he deplores such
other uses of language; nor will he admit that the study of
them has anything to do with philosophy. The business of the
philosopher, as of the scientist, is exclusively with truth, and
the salvation that it offers is that which lies in knowledge of
the truth about the nature of things. This is why Russell
excludes ethics from philosophy: ethics, as he believes, gives
us no knowledge of what exists; it is why he is not interested
in aesthetics: art is not a form of knowledge; and it is why he
hates traditional theology and religion: they give us no un-
derstanding of the nature of things, but on the contrary visit
us with illusions which obscure our vision of what there is. In
short, no activity is worthy of the name of "philosophy" un-
less, at the least, it is a substitute for vision; and science is the
god of thought because and the philosophies that emulate
it,
it, alone provide a reliable substitute for and guide to vision.
The professed aims of Bertrand Russell and Earle's traditional

philosopher are thus very close; their differences are differ-
ences only of idiom, detail, and perspective. And I am sure
that, by and would read
large, the traditional philosopher
Russell's History of Western Philosophy with fascination and
sad approval. "So this," he would say, "is what my vision
comes to when it is corrected by the glasses which have been
ground for me with the tools of twentieth-century analytical
philosophy. The world looks smaller, in a way, but also much
clearer. So be it."
Is this what Mr. Earle himself wants? I cannot believe it. I
cannot believe that he thinks, like the gnostic, that vision is

all, or even the all of philosophy. I cannot believe that like
other gnostics, from Plato to Russell, he thinks that the aim

of the spiritual life is vision, either literally or metaphorically,
by description. Nor can I believe
either directly or indirectly,
that he thinks that every other cultural activity is subordi-
nate to, if indeed it is not an inferior form of, knowledge,
however "knowledge" may be defined. But if I am right, then
who, really, are his friends and allies? Or better, since it is
culture we are talking about, who, among contemporary phi-
losophers, are the true friends of culture? Not, certainly, the
time-serving culture-vultures who batten on the carrion of
the dead tradition; not the phenomenologists whose analyti-
cal and visionary obsessions are devoid of saving grace; and
not the existentialists who, for all their concern, are devoid
both of reason and of lucidity. Who, then? I submit that they

are the much abused philosophers of ordinary language
( many of whom write with great wit and charm as well as
with clarity ) who, following the leads of Berkeley, Kant, and

Wittgenstein, have enabled us at last to understand in detail
how the forms of discourse which serve as leading parts to
art, morals, and religion may be
the spiritual activities of
fully lucidand in their own way rational even though they
describe no matters of fact, provide us with no clear and
distinct ideas of the nature of things, and, in short, tell us
nothing about the world around us.
To the casual observer, such slogans of the ordinary-
language philosophers as "language should not be treated as
a calculus," "notall words function as names or signs of ob-
jects," "themeanings of words are to be found in their uses,"

may not appear world-shaking. But it is through their ap-
plication in the study of moral, religious, and critical ideas
that we have begun to see how profoundly our basic cultural
activities have been misunderstood and subverted by phi-
losophers throughout Western history. At long last we may
understand, for example, why the only alternatives to the
traditional view of art as imitation need not be those anti-
intellectualistic theorieswhich regard art either as an ar-

rangement of merely sensuous surfaces or as an expression
of pure emotion. Similarly, we begin to see why the only
alternative to a supposedly scientific ethics which treats
moral principles as laws of nature is not an ethics of pure
decision which denies any place to reason in moral delibera-
tion and criticism. Morality, like art, provides, in context, its
own meaningful standards of propriety, validity, and reason-
ableness. In use, moral principles function neither as arbi-
trary expressions of emotion which cannot be criticized
without begging the question nor as scientific laws which de-
scribe or predict certain supposedly objective features of the
moral universe. But there is no way to see this without de-
tailed study of what we actually do with words when we
state a moral principle, apply it to a particular case, or dis-
pute its same way we may come to under-
validity. In the
stand, in the end, how theology and even metaphysics itself
may have significance even though what they may tell us has
no validity as an explanation either of mundane or of super-
mundane realities. The problem, again, is to discern what the
theologian and metaphysician are doing with language. It
has been said that the God of the philosophers is no God at
all but a bogus principle of explanation or "ground of being."
But, if so, this is primarily because traditional philosophers,
ignoring the contexts of its ordinary religious use, have mis-
conceived the meaning of "God" in treating it as a term of

transcendent reference. How little such a conception of God
has to do with actual religious thought, however, is only just
now becoming clear through the analytical study of our talk
of God in the familiar circumstances of religious prayer and
meditation.
Thus, by directing our attention to the various ways of
words in their different authentic vocations, and to the tradi-
tional human practices with which they are commonly cor-
related, the much-abused philosophers of ordinary language
are bringing us gradually to recognize that in any sphere
lucidity is as lucidity does, that "reason" is not the sole
prerogative of the scientist and the logician, and that even
what we call "knowledge" need not be assumed in advance

to be a representation of what traditional philosophers under-
stood, or thought they understood, by "existence." In a way,
it may turn out that through a more liberal and relaxed use
of the methods of analysis employed by the ordinary-language

philosophers even the strange-sounding talk of the existen-
tialists may be seen for what it is, not as an obscure or ir-
rational description of what "exists" in return, natura, but as

a faltering but intelligible effort on the part of care-ridden
and anxious men to tell each other what it means to "face
reality" without the illusion of a Logos.
LOGICAL ANALYSIS AND THE LOVE OF WISDOM

It is now necessary to place these remarks in a somewhat
different perspective. I have argued that until very recently
technical analytical philosophy itself has belonged in the
direct line of Mr. Earle's gnostic tradition. However, there is,
one salient feature of that tradition which is conspicuously
absent from contemporary technical philosophy of nearly all
types. This is the notion that there is a unique philosophical

mode by Spinoza, for exam-
of salvation of the sort depicted
ple, in the Fifth Part of his Ethics.Mr. Earle regrets it, but I
am bound to say that, to my mind, this is good riddance.
The "visions" of the traditional gnostics and rationalists were
not synoptic but pseudo-visions. There is no esoteric philo-
sophical gradus ad parnassum through which we may be led
to a saving glimpse of "the Good" that lies, supposedly, at
the core of being itself. And if such be "wisdom," then we
are better off without the spurious comforts which it affords.
As Socrates knew, but Plato and Spinoza forgot, the philoso-
pher is not by vocation a kind of secular savior; at his best,
he is a midwife whose first function is to clarify and connect
the concepts by which we give focus to the ends that guide
us in the ordinary conduct of life. Philosophy, in short, pro-
vides no access to a unique reality which may console us, in
Mr. Earle's words, for our "wrecked lives." On the contrary,
to attempt as the gnostics do, to make philosophy serve as a
natural or lay religion is to pervert its primordial and proper

Socratic function. It is also to destroy religion itself.
This is not to deny to the philosopher his ancient claim to
be a lover of wisdom. Rather it requires a very different view
of the wisdom that he seeks or may command. Now I am
aware that the enigmatic commandment of the Oracle,
"Know thyself," has been given many interpretations in the
past. But I believe that Socrates' interpretation of it, properly
construed, is the correct one. I also believe that Socrates and
not Plato more nearly understood what contribution philo-
sophical reflection may make to our self-knowledge and self-
culture. That contribution is essentially the logical (or dia-

lectical) analysis of the golden concepts, such as justice,
truth, God, through which we focus the ideals by which we
live. In a spiritual sense, we are our commitments. Therefore,
if these concepts provide the terms through which alone our
commitments are made manifest to ourselves and others, then

their careful study must provide a revelation of what we are,
both to ourselves and to one another. Stated in another way,
philosophical analysis is, at bottom, nothing but that device
by means of which we attempt to discover, or uncover, what
we really are. And I use the word "discover" here advisedly.
We have been made to think both by the existentialists and
by the positivists that our basic commitments and ends are
fundamentally a matter of decision or choice. Nothing could
be further from the truth. I do not choose what I mean, or
intend, by truth or by justice; nor do I decide what princi-
ples of knowledge or of justice I will live by. I find myself
talking and thinking in a certain way just as I find, some-
times with great difficulty, those ideals and procedures by
which I live.
If this is so, then, making due allowances for understand-

able shifts in perspective and for inescapable changes in
philosophical style, it no longer seems far-fetched, despite
Mr. Earle, to regard the most hard-bitten positivist as still

involved, in his own quaint way, in the ancient Socratic
quest for self-knowledge. In spite of himself he remains, in
the classical sense, a philosopher. Even the most tough-
minded exponents of "analysis" in philosophy are still not

interested merely in empirical descriptions of our linguistic
behavior. What concerns them, one and all, is what they
variously call the "clarification," the "logically correct" or
"proper" uses of expressions. And what they want to know is
not what we do say, but what we ought to say or would
say if we
spoke well or to some purpose; or if they are inter-
ested inwhat we do say, then it is on the assumption that
proved ways of speaking are best and that when we ignore
them, as the traditional philosophers so often did, we go off
the rails and so defeat our own ends. Nor are technical
philosophers equally interested in all of the terms and sen-
tences that ordinary men are prone to utter. The terms in
which they are mainly interested, such as "knowledge,"
"truth," "validity," and "meaning," are not used in order to
designate items in the world of phenomena; they are, on the
contrary, normative expressions that are employed in guid-
ing and connecting the and otherwise,
activities, linguistic
through which men who live in communities organize their
lives. It is the procedure which concerns the philosopher, not
the fact; the method, not the particular results that may be
achieved through its application.
But the matter goes deeper than this. The older philoso-
phers could regard words as mere counters precisely because
they viewed both language and thought as finished photo-
graphic representations of an independent, abstract, and
equally static reality. But if we take time and action seri-
ously, then it becomes apparent that language simply does
not exist save "performatively" in the living speech-acts in
which men do the various specialized things we call "com-
mending," "promising," "advising," "inferring," and "describ-
ing." In a word, language really exists only in act, so that if
we would grasp the meanings of words ( or ideas ) we must ,
also grasp what, in particular, men are doing with them. And
conversely, when we do manage to understand what expres-
sions mean, we must necessarily also understand to that ex-
tent what the men who use them are up to, what they are
doing, intending, and aiming at. And it is partly for this
reason that Wittgenstein was able to say with such profound

truth that forms of words are also forms of life.
Thus, although language does not provide a mirror of the
external world, as the classical philosophers supposed, it may,
when properly understood, serve as a revelation of ourselves.
That revelation may disappoint or even disgust us. In that
case, both the forms of words and the forms of life which
they reflect will appear as something to be transcended. Or
it may disclose aspects of our existence which, as we thought-
fully contemplate them, are found to be immensely precious
and which must be defended at all costs against those who
would, witlessly, destroy them. Only from this standpoint
may we begin to comprehend the underlying spiritual signifi-
cance of the closet-quarrel between the ordinary-language
philosophers and their opponents. On the one hand, the
proponents of ideal scientific languages, the logical revision-
ists and reconstructionists, the rule-makers and symbol-
mongers are, in effect, revolutionaries. At bottom they are

dissatisfied not onlywith existing linguistic practices but also
with the lives which are reflected in them; and they would,
if they could, refashion their own lives and societies in ac-
cordance with some ideal image toward which they, perhaps

hopelessly, yearn. Or, they are romantic rebels who, like Mr.
Earle himself, dream of a land beyond the seas where new
men may start from scratch to build the true Republic of
Knowledge of which they already think they have a glimpse.
On the other hand, the philosophers of ordinary language,
as I envisage them, are classicists and traditionalists who,
without illusions of perfection, seek to understand, and
through that understanding to keep alive, the forms of lan-
guage and of life which, partly by instinct and partly by
conviction, determine what we are. Philosophically, the lin-
guistic studies of these philosophers are revolutionary. Hu-
manly and however, their effect, if not also their
spiritually,
intention, is to conserve practices, including linguistic usages,
which may be threatened by our increasingly technological
and scientific culture, but which are nonetheless essential to
the survival in any form of spiritual concerns whose ends are

not definable in purely "cognitive" terms.
The fundamental tragedy of rationalism in philosophy is
that it forces its victims to misconceive their own cultural

intentions. If they love art, then they must justify this love by
treating the symbolic forms of art as modes of representation,
in which case they cut themselves off, as Mr. Earle has done,
from artistic styles which, in no obvious way at least, imitate
nature. If they love the good, then they must needs justify
this by representing goodness as an objective quality or rela-
tion in things, thereby risking the chance of moral cynicism
when they discover, as they must, that no such quality or
relation exists. If they love God, they must conceive him as a
glorified "substance" or "cause," for the only form of "being"
their philosophies permit them to acknowledge is the being
of an object of knowledge, thus sacrificing the possibility of
religious worship on a misbegotten altar of cognition. Worse
than this, they are also obliged to misrepresent those who,
like Augustine, Pascal, Hume, and Kierkegaard, have spo-
radically defended an a-gnostic and pluralistic conception of
man's spiritual well-being. The latter are treated, in brief,
not as a loyal opposition which merely rejects a certain
philosophical conception of human culture, but as irrational-
istsand mystagogues who, like the Grand Inquisitor, would
replace culture with an authoritarian cultus of mystery,
miracle, and indoctrination. In fact, however, it is the off-
beat a-gnostic rebels who have more truly understood the
complex spiritual intentions of Western man and who have
therefore better grasped the kinds of lucidity that are proper
to each. And it is the philosophers of ordinary language who,
by the slow but sure process of piecemeal logical analysis,
are beginning to spell out the underlying meaning of those
intentions.
I am anxious not to be misunderstood; I completely agree
with Mr. Earle that the contempt of reason is a form of

"blasphemy." But blasphemy has many guises, and perhaps
the most insidious of all is that strange hypertrophy of reason
which, since the time of Plato, has always been the unwitting
parent of irrationalism. In our time, the same error is re-
sponsible for that romantic and ultimately barbaric mysology
which currently infects so much of our contemporary the-
ology, criticism, and anti-analytical philosophy. Reason, it
cannot be too often said, is not a Logos, but an activity of

the human mind. In the course of history, it has taken many
different forms and its standards have been constantly sub-
ject to change in accordance with man's changing conception
of himself. We cannot possibly rehabilitate the gnostic cult of
reason which dominated Western philosophy until the end of
the eighteenth century. And we cannot do so because, since
Kant, the devastating criticism of that cult has undermined
the spurious mystique upon which it rests.
CONCLUSION
In conclusion, let us return to Mr. Earle's own analysis of
culture and his dreams for its new life.
In my opinion, the task of revitalizing our culture, if that is

what it needs, must be conceived in very different terms from
those which Mr. Earle employs. The metaphors of birth, life,
and death are especially misleading when applied to com-
munal cultural activities such as art, religion, and philosophy.
For one thing, they result in stances which engender a sense
of fatality and helplessness in the presence of an irrevocable
round of vital processes which human thought and choice
are powerless to affect. Or
encourage that revolu-
else they
tionary desperation which causes men to run amok, wantonly
destroying anything that reminds them of a golden youth of
culture which is beyond recall. Does Mr. Earle fully realize
how ambiguous is his own sense of direction at this fatal
crossroad between the old regime and the terror?
In one way I make no complaint. Wise men, wherever they
are, always live ambiguously in the middle way between
traditionand revolution. The part of intelligence is to con-
serve and reform the old patterns when we can and to
fashion new ones, gradually, as we must, creating but also
re-creating our spiritual destinies, piecemeal, as we go along.

In some sense, no doubt, God, Nous, the work of art, and
hence culture itself are "dying." It may be a fact that
culture in Mr. Earle's sense is actually dead, once and for
all. In that case, however, our problem is to learn to live
without it. Is this prospect so bad? If we tried, we might

find that we no longer really want that absolute certainty,
ultimate lucidity, and communal identity to which our an-
cestors aspired. It has been said that our responsibilities be-
gin in dreams. But I, for my part, am able to dream of a
society (not, in Mr. Earle's sense, a community) of emanci-
pated, civilized men who, through reticence, tact, and self-
control, are able to redeem the time through consecutive
acts of creation, intelligence, and love, but without insisting
that each of these be viewed as part of some great communal
rite. Such men, as I fancy them, would refuse to be over-
borne by the thought of death, either for themselves or for

their civilization. On the contrary, they would find in the
amenities and in the work of civilized life an honest alterna-
tive to the morbid and self-destructive preoccupation with
"ultimate concern," whose objectless ends they cannot, fi-
nally, fathom. They would acknowledge the wondrous mys-
tery of being without making a fetish of that wonder, and
they would accept, without pride, the daily satisfactions that
may accrue in the ordinary course from the performance of
the specialized tasks which constitute their "callings." Such a
dream, I realize, is very uncatholic, just as it is also very un-
Hegelian and un-Marxian. So be it. If it implies, as I do not
for a moment doubt, that we must in one sense live "de-
spairingly," then we may learn to make the most of that too.
Here the philosophers have much to learn from the great
tragic poets who have shown us, through their art, how de-
spair itself may be redeemed when it is finally accepted, and
thus transmuted, as a condition of one's being.
I realize how much against the grain such a question may
go, but it must be asked: Is not the very notion of a "whole
man" itself a tragic myth, a philosophical bait unwittingly
laid to make the taker mad? Organic analogies are usually
acknowledged to be a mistake when applied on the social

level;perhaps the mistake occurs also when such analogies
are applied to the spiritual "life" of an individual person. I
suggest that Mr. Earle, like all gnostics, past and present,
has been victimized by the erroneous idea that the spirit or
the self is an organic substance. In short, Mr. Earle's vision
of culture is a bogus vision because the philosophy behind it
is bad philosophy. Why should we aspire to belong to one

great communal "substance" when the evidence of our senses,
our hearts, and our discourse informs us that spiritual diver-
sity is as irreducible, as ultimate, and as certain as death
itself? But even if, in some sense, we all were "one," what
possible spiritual comfort or nourishment could be derived

from that wretched fact? What if the one were a devil, a
dictator, or a rock? And if it comes to that, why, by analogy,
may it not be so?
At the human level, autonomy is the prerogative of all self-
respecting men. It is also the prerogative of our various con-
cerns, each of which, so long as it does not crowd the rest,
has its inviolable right to be. The spiritual life requires craft,
technique, and therefore specialization. Why, then, should
we deny it. A
good poem is not an analysis of concepts, and
talents which make for good poetry often merely come to
grief when turned to the chores of philosophical midwifery.
An arrangement of "meaningless" sounds in a satisfying aural
pattern is not everything, but done by a master like Mozart
it may serve to delight the spirit even of a Plotinus. We are
not and cannot always be doing or caring about one big

thing, and, thank God, we are not always doing our various
little things together. Nor does doing what we do with "our
whole heart" require that it should be all in all to us. Mr.

Earle's "work of culture," as I call it, is not a seamless web
of holy, good, and beautiful Truth, but a sentimental ab-
straction to which nothing in modern life corresponds. Its
only conceivable analogue is a social institution, viewed in
Hegel's sense as an evolving historical form of "objective
spirit." But Hegel himself knew, as Earle seems to have for-
gotten, that art, philosophy, and religion belong to the do-

main of "absolute spirit" within which each individual must

work out his own unique destiny according to his lights.
The fundamental difficulty with Mr. Earle's dream comes
back to the matter of language. I am sure his talk of "spirit"
as something which is "alive only when creating itself and
when concerned "precisely for itself, for its life," is only a
manner of speaking. But it is a manner of speaking which is
a philosophical booby-trap. As he himself says, at one point,
"the human spirit is inevitably in individuals." Concretely
this means that it has no "being" and no "substance" what-
ever, save in the particular activities, tasks, and aspirations
which occupy individual men during their ordinary waking
hours. We attain to spirituality not when we dream, but only
when we try to interpret the meaning of our dreams. As
Santayana put it, what we call "spirituality" is mainly a
matter of wakefulness. It neither lives nor dies; it has no
independent concerns, but attaches itself, willy-nilly, to any
concern, ultimate or otherwise, as we become conscious of
its end. It has no ideas; but any idea, when understood, be-
comes a spiritual thing; and it can never express itself, for it
has nothing in itself to say or tell. It is neither cultured nor

uncultured, civilized nor uncivilized, good or bad. Slaves and
freemen, saints and sinners, nominalists and realists, gnostics
and pistics, and paper hangers may all

portrait painters
possess it, each in his own way. That it has its troubles in
the modern world I should be the last to deny. But its
"death" is not due to specialization; on the contrary, spe-
cialization, like attention, is its life. Its death is due to dis-
traction, sleepiness, boredom, and anxiety. And for these
there is no single remedy.
.
Index
Abraham, 171, 224 77; of moral rules, 84; pre-

Aeschylus, 261 sumptive principles of moral,
Alciphron, 359
xi, 141ft.
Alexander, 315
S., Ayer, A.J., 36, 367*, 451*, 113
Allport, Gordon, 181
Analysis, Philosophical, and the
Spiritual Life, 34gff Baier, Kurt, 168
Analytic ethics, 11, 24-32 Bentham, Jeremy, 55, 243, 297,
Analytical philosophy, 5, 11 354
Antidescriptivism, 45ff. Berenson, Bernard, 319
Antigone, 80, 142, 260 Bergson, Henri, ix, 315
Antinomy of moral objectivity Berkeley, George, xi, xii, 238,
and freedom, i3gff. (Bishop of Cloyne) 355,
Aquinas, Thomas, 7311, 176, 358, 361, 365
242 Black, Max, ign, 211
Aristotle, 5, 29, 88, 242, 257, Bradley, F.H., 48, 83, 224
265, 287, 359 Brentano, F., 232
Arnold, Matthew, 350 Broad, C.D., xii, 12, 54, 82
Auden, W.H., 351 Buber, Martin, 185, 352
Augustine, 171, 173, 205, 371 Burke, Edmund, 241
Aurelius, Marcus, 31 Butler, Joseph, 305
Austin, John, xii, 180, 238
Authority of moral judgments,
11 1-27 Carnap, Rudolf, vjn, 36, 112,
Autonomy : of ethical discourse, 113,360
.
Index
Cassirer, Ernst, 323 Earle, William, xv, 349ff.

Catholicism, 122 Education, uses of moral phi-
Ceremonial use of language in losophy in, 3-32
ethics, 116-21, limitations Egoism, 59
of, 12 iff. Emotive meaning, 53
Characteristics: "empirical," Emotive theory of ethics, isff.,
14; "ethical," 54; "good-mak- 66, giff., 112-14, 121, 248;

ing," 254; "natural," 14; criticism of, 114-21
"non-natural," 15, 33-4 Emotivists, 22, 71
Claims, competing, ioiff., Epicurus, 31, 52
26off. "Ethical," meaning of com-
Cognitive theory, 66 pared with "moral," 222
Comte, Auguste, 197 Ethical naturalism, revival of,
Conscience, ultimacy of, 27 24off.
Consistency, principle of, 102, Ethical realism, 227-39
202 Ethical thesis, the, I75ff
Conventions, global, in ethics, Ethics: analytical, 7-9, 24-32;
128-33 commonsense, 209-26; sub-

stantive, 6-8, 28
Evil: problem of, I7iff.; solu-
tions of problems of, igoff.

Ewing, A.C., 34-5, 232
Definition: descriptive, 48;
Existentialism, 4
ructions of, 46ff.; persuasive,
Existentialists, xiv, 86
53, 125; practical, 53, 54
Descartes, Rene, 150
Descriptivism, 11, i4ff.
Faith, i73ff.
Determinism and social free-
Feibleman, James, 61
dom, 3o8ff.
Feigl, Herbert, 73?*
Dewey, John, xii, 5, 12, 13, 35,
Fichte, Johann Gottlieb, 285,
36, 111, 113, 197, 247, 318,
324
355 Firth, Roderick, i52ff.
Disinterested observer, i53ff. Flew, A.G.N., 172
Disinterestedness, I53ff.; def- Fosdick, Dorothy, 293
inition of, 154 Frankena, William, xv, 33, 33n,
Dostoyevski, Fyodor, 171, 173, 34-5, 46n, 60
174, 189 Freud, Sigmund, 29, 243, 289
Duty: conflict of duties, 102-5 Fromm, Erich, 298
Index 111
Gnostics, 227, 357ft. Imperative: categorical, 82;

God: and evil, 171-205; con- theories of ethics, 214
cept of, 1766°. ; of the phi- Instrumentalists, 71
losophers, 366 Intuitionism, 33ft.
Goethe, Johann von, 332 Intuitionists, 12, 13, 15, 279ft.
"Good," meaning of, 8ff., 34,

46, 89E
James, Henry, xii
James, William, xii, xiii, xv,

Hall, E.W., 227ft. 238, 331, 346, 355
Hampshire, Stuart, 55, 64ft, 88, Jefferson, Thomas, 291
257 Jesus, 31, 184, 321, 331
Hare, R.M., 188 Joad, C.E.M., 114
Hayek, Friedrich von, 266 Job, 80, 160, 171, 173
Hedonism, 48-51, 55-8, 187, Johnson, Samuel, 88
244-5, 255 Jung, Carl, 243
Hegel, G.W.F., xiv, 80, 203, Justice, principles of, 79, 80
260, 265, 273, 284ft., 316,
324, 35off.
Heidegger, Martin, 315
Historicism: and ethics, 274ft.; Kant, Immanuel, xiii, 5, 17, 27,
critique of, 267ft. 31, 81-3, 88, 105, 139, 152,
History, laws of, 269ft. 238, 250, 253, 311, 333, 357,
History, Morals, and the Open 361, 365, 372
263—92
Society, Kaplan, Abraham, 132—3
Hobbes, Thomas, 31, 305, 359 Kierkegaard, Soren, xiv, 185,
Holmes, Oliver Wendell, 302-3 199, 219, 224, 346, 350,
Hume, David, xii, xiii, xv, 28, 356, 362, 371
59, 71, 78, 79, 88, 129-30,

152ft., 175, 238, 243, 247,
306ft., 322, 357, 371
Language: and ontology, 233,
Husserl, Edmund, 356
362ft.; of conduct, 21-3, 24,
Hutcheson, Francis, 152
39, 66-7; philosophy of or-
dinary, 209ft., 365ft.; plural
uses of, 19, 33ft.; roles of the
Idealists, 17 language of conduct, 33-43
Ideal observer, 152ft. Lasswell, H.D., 132-3
.
IV Index
Law and morals, 276ft. 46-8, 90, 146, 210-11, 227-

Leibniz, Gottfried von, 192, 8, 244-5, 289, 338, 363
357, 359 Moral: deliberation, 71, 72;
Levels of Moral Discourse, freedom, i39ff.; nature of,
65-87 i6iff.; objectivity, 134—70;

Lewis, C.I., 29, 34, 203, 255 philosophy and education,
Liberty: moral foundations of, 3-32; practices, 164; pre-
31 off.; utilitarianism and, sumptive principles of, 139-
293-314 41; principles, 27, 102-5,
Linguistic analysis, 9 140; reasoning, 7, 71, 88-
"Linguistic philosophy," xi, xiii 110; rules, 74, 81, 98; truth,
Lippman, Walter, 242—3 i67ff., 333ff.
Locke, John, 5, 171, 242, "Moral," meaning(s) of, 22off.
303-4 Mozart, W.A.

Logical analysis and the love of
wisdom, 367ff.
Lucretius, 330
Naturalism, ethical, 145, 24off.
Naturalistic fallacy, 44ft.
Nicomachean Ethics, 29
MacDonald, Margaret, 1 14,
Nietzsche, Freidrich, xiv, 144,
n6ff.
145, 173, 215, 226, 289,
Malinowski, B., 72, 117
35o, 353, 357, 360
Matisse, H., 337
Northrop, F.S.C., 329
Marx, Karl, 243, 265, 273, 289,
290, 291, 350
Mayo, Bernard, 167-8
Meaning: descriptive, 16, 18, Objectivity, moral, 134-70
19; "emotive" distinguished Obligation, 12
from "cognitive," 18 Occam, xi
Metaphysics, viii Ogden, C.K., 15, 18
Mill, James, 312 "On Social Freedom," 293ft
Mill, John Stuart, 31, 48, 49- Open question argument, 14,
52> 54-5, 57-9, 79, 81, 188, 145, 146
225, 243, 246, 253, 273, Open society, History, Morals,
293-3M, 353ff- and the, 263—92

Monotheism, 171ft. Ordinary language: common-
Monotheistic "syndrome," i86ff. sense ethics and, 209-26;
Moore, G.E., 8, 12, i3ff., 35, philosophy of, x, 19—20
.
Index
'Ought": the ethical, 7, 10, 37, Prior, A.N., 44, 44TI, 59

50, 66, 77, 80; use of, 109 Pritchard, H.A., 35, 70, 71
Promise keeping, 98-100
Protestantism, religious, 122-4
Pascal, B., 188, 245, 371

Pater, Walter, 319
Question: "open," 14, 145, 146;
Peirce, C.S., xii, xiii, xv, 238,
"limiting," 209ff., 224
355
Quine, W.V., xv
Pepper, Stephen, 333, 334,
337, 338
Perry, Ralph Barton, xvi, 12,
35, 35", 36, 47, 63, 177, Rationality, Santayana's inter-

23off ., 247, 262, 334, 336 pretation of, 323s
Person: concept of a, 18 iff.; Rawls, John, 72-3
"holy," 184; "moral," 184 Realism, ethical, a defense of,
Philosophical analysis: and the 227-39

Spiritual Life, 349ft. ; schools Reason, deductive, 94ft.
of, 21 off. "Reasonable," meaning of,
Philosophy: analytical, 5; i2gff.
"first," viii Reasoning, moral, 7off., 88-110

Plato, 5, 52-3, 89, 135, 243, Rees, J.C., 293E
264ff., 314, 316, 317, 355, Relativism, Protagorean, 25
358-9, 365, 367ff- Religion and ethics, i7off.,
Pleasure, 255ff. 225ft.

Plotinus, 180, 357, 374 Rice, Philip Blair, i28n, 24off.
Popper, Karl, ix, ssn, 56, 114, Richards, LA., 15, i6n, 18,
263-92 112, 209
Positivism, "logical," 4, 16, 210 "Right," meaning of, 7, 12
Practices: as commitments, Robinson, Richard, 52-3, 53n
164; conflicting, 165 Ross, W.D., 12
Pragmatism, 34 Rousseau, Jean Jacques, 245
Pragmatist, xii Royce, Josiah, 351
Prall, David, 329 Russell, Bertrand, xi, xii, 5, 20,
Principles: ethical, 75, 81; of 210, 317, 318, 322, 331,
moral autonomy, viiff., 338ff., 352, 355, 359, 36off.
14 iff.; of moral objectivity, Ryle, Gilbert, 58, 210, 211,

I39ff- 281, 361
VI Index
Santayana, George, xiv, xv, 33ft; "emotivism," isff., 33,

315-48, 356 66, goff., 112, 121, 248; im-
Science and history, 2696:. perativism, 214; intuition-
Scientific philosophy, x, xi ism, 11, 33, goff., 212, 245,

Selby-Bigge, L.A., 55*1 279; misleading classification
Shakespeare, William, 260 of, 54ff.; naturalism, 11, 33,
Sidgewick, H., 90 90, 111, 212, 240
Skinner, B.F., 101—2n Thoreau, Henry David, 223
Smith, Adam, 152 Tillich, Paul, 201, 352
Social behavior, organization Toulmin, Stephen, ig-2on, 72,
of, 21 73, 88, 93, 96, 2i2ff., 229,
Social Freedom, On, 293ft 257, 314
Socialism, 20,off. Toynbee, Arnold, 273
Society: "open," 80, 263-92; Tragedy, 259-61
"closed," 72, 2655.
Socrates, xiv, 242, 317, 324, Utilitarianism, 57, 100, 261;
345-6, 367ft and justice, 100; and liberty,
Spencer, Herbert, 316
293-314
Spengler, Oswald, 273
Utility, principle of, 297ft
Spinoza, Baruch, xiii, 189, 316,
317, 335, 355ft
Validity, 62, 106, 109
Stevenson, C.S., xv, i8n, 34,
"Value," 7, 12; judgments, 18,
52-3n, 62-3, 71, 92, 93,
26; what is it?, 227ff.
106, 113, 114, 121, 124-5
Veblen, Thorstein, 263, 264
Symbolic forms: George San-
Voluntarism, 57
tayana, natural historian of,
Voluntarists, 56
3i5ff., George Santay ana's
contributions to the study of,
320ft White, Morton G., xv, 60, 270,
355, 356
Whitehead, Alfred North, 315,
Theodicy, ig4ff. 322, 355, 366
Theological thesis, the, i75ff.; Wild, John, 264-5
meaning of, 176-9; moral Wisdom, John, vii, 210
claim of, 179-85; religious Wittgenstein, Ludwig, x, xii, 5,
claim of, 185-6 ign, 210-11, 228, 233, 234,

Theories of ethics: ceremonial, 238, 241, 352, 361, 365, 370
116, 12 iff.; classifications of, Working model, 1 146:.
A NOTE ON THE TYPE
The text of this book is set in Caledonia, a Lino-

type face designed by W. A. Dwiggins ( 1880-
!956), the man responsible for so much that is
good in contemporary book design and typography.

Caledonia belongs to the family of printing types
called "modern face" by printers —a term used to
mark the change in style of type-letters that oc-
curred about 1800. Caledonia borders on the gen-

eral design of Scotch Modern but is more freely
drawn than that letter.
Composed, printed, and bound by

H. Wolff, New York.
Typography and binding design by
VINCENT TORRE
Due Reti"
i
Reason and conduct
I It
main
171A291rC2
Win.
e .A
3 12b3 03307 OfiSM
KEEP CARD IN POCKET
IT IMPORTANT THAT
IS
CARD BE KEPT IN POCKET

H. D. Aiken 1961 - Reason and Conduct

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

H. D. Aiken 1961 - Reason and Conduct

Uploaded by

Copyright:

Available Formats

UNIVERSITY

& & &

HENRY DAVID AIKEN

NEW YORK : ALFRED -A KNOPF

L. C. catalog card number: 62-12564

THIS IS A BORZOI BOOK,

Although the papers selected for inclusion in this volume have

a group of problems that appear to me crucial for moral phi-

these essays, not some others, that I have thought worthy, if

fact,which had temporarily been forgotten, that morality

but also, in turn, determined by the idea or ideal of them-

removable from our whole conception of the moral life. In

affinities and to define their limits. That I owe much to the

English philosophers of ordinary language will be apparent

ever sought to preserve a continuity between philosophical

tists as such, and that only when such (unknown) translations

have been made is it possible to determine the truth-value of

philosophy to Harvard students, both graduate and under-

celebrated by the followers of Wittgenstein. Something of the

is exhibited throughout his Treatise with results that are in-

wrote my doctoral dissertation, prepared me well for the ethi-

never occurred to me that the study of language could be

From the pragmatists I learned also that what men call

before the age of the historical consciousness, and it remained

"the improvement of the understanding" (or reason) is not

and a variable sentiment, whose objective forms are in many

critique. This error lies at the root of the notorious "panlog-

long way round to Nirvana," and that in the course of trying

self, whose philosophy James once described as "the perfec-

which the life of reason itself is always incysted. And in that

than to those of any other single philosopher save Hume.

possibly realize. To have such students is itself sufficient com-

without whose encouragement the idea of gathering these

Henry David Aiken

vi • The Authority of Moral Judgments 111

vn • Global Conventions in Ethics 128

vin • The Concept of Moral Objectivity 134

xm • History, Morals, and the Open Society 263

Index follows page 375

CURRENT PHILOSOPHY AND LIBERAL EDUCATION

likely is any new development in a branch of learning to

revival of educational interest in philosophical problems is

occurring precisely at a time in which there has been a

both of educators and of the general public that a sound

result has been that philosophical theories of great rele-

a difference, it lies rather in the fact that contemporary an-

SUBSTANTIVE ETHICS AND ANALYTIC ETHICS

Most analytical philosophers are now agreed that a primary-

the general failure to distinguish between two very different

self to the question how we ought to live as "substantive

bility of substantive moral philosophy as a discipline pre-

Analytic Ethics. The questions of substantive ethics are

persons. Questions of this sort are the following: "What are

1 Speculation is not opposed to practice. Nor is it always directed to

mind at the outset, however, that interest in analytical

of a substantive "moral science." How dubious such a sup-

sential to understanding of the tasks of analytical ethics to see

asking about the meanings or uses of words. They are ob-

ment and hence seriously to hope for a settlement of the

3 I.e., whether there is in fact some objective quality or relation

sight problems of meaning may appear to have little rele-

bread." On this point not merely the opponents of analytical

THE DEVELOPMENT OF RECENT ANALYTIC ETHICS